Dissertations / Theses on the topic 'Intelligent surveillance system'
To see the other types of publications on this topic, follow the link: Intelligent surveillance system.
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 dissertations / theses for your research on the topic 'Intelligent surveillance system.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse dissertations / theses on a wide variety of disciplines and organise your bibliography correctly.
1
Zhou, Han, and 周晗. "Intelligent video surveillance in a calibrated multi-camera system." Thesis, The University of Hong Kong (Pokfulam, Hong Kong), 2011. http://hub.hku.hk/bib/B45989217.
Full text
2
Charvat, Robert C. "Surveillance for Intelligent Emergency Response Robotic Aircraft (SIERRA Project)." University of Cincinnati / OhioLINK, 2012. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1337888115.
Full text
3
Du, Ruixiang. "An Intelligent Portable Aerial Surveillance System: Modeling and Image Stitching." Digital WPI, 2013. https://digitalcommons.wpi.edu/etd-theses/859.
Full textAbstract:
"Unmanned Aerial Vehicles (UAVs) have been widely used in modern warfare for surveillance, reconnaissance and even attack missions. They can provide valuable battlefield information and accomplish dangerous tasks with minimal risk of loss of lives and personal injuries. However, existing UAV systems are far from perfect to meet all possible situations. One of the most notable situations is the support for individual troops. Besides the incapability to always provide images in desired resolution, currently available systems are either too expensive for large-scale deployment or too heavy and complex for a single solder. Intelligent Portable Aerial Surveillance System (IPASS), sponsored by the Air Force Research Laboratory (AFRL), is aimed at developing a low-cost, light-weight unmanned aerial vehicle that can provide sufficient battlefield intelligence for individual troops. The main contributions of this thesis are two-fold (1) the development and verification of a model-based flight simulation for the aircraft, (2) comparison of image stitching techniques to provide a comprehensive aerial surveillance information from multiple vision. To assist with the design and control of the aircraft, dynamical models are established at different complexity levels. Simulations with these models are implemented in Matlab to study the dynamical characteristics of the aircraft. Aerial images acquired from the three onboard cameras are processed after getting the flying platform built. How a particular image is formed from a camera and the general pipeline of the feature-based image stitching method are first introduced in the thesis. To better satisfy the needs of this application, a homography-based stitching method is studied. This method can greatly reduce computation time with very little compromise in the quality of the panorama, which makes real-time video display of the surroundings on the ground station possible. By implementing both of the methods for image stitching using OpenCV, a quantitative comparison in the performance is accomplished."
4
Barake, Bassem. "Towards an intelligent surveillance system for public security at crowded places." Thesis, University of Ottawa (Canada), 2009. http://hdl.handle.net/10393/28258.
Full textAbstract:
Nowadays, safety and security of public areas has become the centre of attention, and especially after 9/11 attack. A visual surveillance for public security is proposed to meet the needs of public areas, such as shopping malls, train stations, airports, etc...
Surveillance researchers from the computer vision have focused on building automated systems and have hardly adopted an approach, where human is involved in decision making with the help of computer inputs. Knowing that the fully automated systems have some advantages such as saving human power, and supporting remote monitoring, but in other hand, these systems are not often tended to work in complex environments. This thesis adopts a human-cum-machine centric approach, where this approach is followed in order to assist security personnel in the physical monitoring with the support of video surveillance. A proposed system is designed and developed, which provides security personnel with real-time information about any suspected person they want to suspect about. Using the security system's application, the officer has to identify the region within the surveilled area where the suspected person is located. Based on the region identified, the system delegates and controls nearest installed cameras to cover the region and then capture pictures of the scene. Once the suspected person is identified by the security personnel from the pictures captured, the corresponding information such as personal and contact information of the person is displayed and available to the security officers. The architecture of the system is based on web services technology, where many researches outlined the benefits of using web services in terms of scalability, reliability, and re-usability of its components. This thesis provides detailed information about the proposed system and its components, discusses the application's graphical user interface to be used by security personnel, and proposes a methodology to select and control installed security cameras to cover a region identified by security personnel.
5
Valera, Espina Maria. "An approach for designing a real-time intelligent distributed surveillance system." Thesis, Kingston University, 2006. http://eprints.kingston.ac.uk/20298/.
Full textAbstract:
The main aim of this PhD is to investigate how a methodology rooted in systems engineering concepts can be established and applied to the design of distributed wide-area visual surveillance systems. Nowadays, the research community in surveillance systems tends to be mostly focused on the computer vision part of these systems, researching and developing more intelligent algorithms. The integration and finally the creation of the system per se, are usually regarded as a secondary priority. We postulate here that until a robust systems-centred, rather than algorithmic-centred approach is used, the realisation of realistic distributed surveillance systems is unlikely to happen. The future generation of surveillance systems can be categorised, from a system engineering point of view, as concurrent, distributed, embedded, real time systems. An important aspect of these systems is the inherent temporal diversity (heterogeneous timing) that arises from a variety of timing requirements and from the parallelisation and distribution of the processes that compose the system. Embedded, real-time systems are often naturally asynchronous. However, the computer vision part of these surveillance systems is commonly conceived and designed in a sequential and synchronous manner, in many cases using an object-oriented approach. Moreover, to cope with the distributed nature of these systems, technologies such as CORBA are applied. Designing processes in a synchronous manner plus the run-time overheads associated with object oriented implementations may cause communication bottlenecks. Perhaps more importantly, it may produce unpredictable behaviour of some components of the system and hence undetermined performance from a system as a whole. Clearly, this is a major problem on surveillance systems that can often be expected to be safety-critical. This research has explored the use of an alternative approach to object-orientation for the design and implementation of intelligent distributed surveillance systems. The approach is known as Real-Time Networks (exemplified by system engineering methodologies such as MASCOT and extensions such as DORIS). This approach is based conceptually on conceiving solutions as being naturally concurrent, from the highest level of abstraction, with concurrent activities communicating through well-defined data-centred mechanisms. The methodology favours a disciplined approach to design, which yields a modular structure that has close correspondence between functional elements in design and constructional elements for system integration. It is such characteristics that we believe will become essential in overcoming the complexities of going from small-scale computer vision prototypes to large-scale working systems. To justify the selection of this methodology, an overview of different software approach methods that may be used for designing wide-area intelligent surveillance systems is given. This is then, narrowed down to a comparison between Real-Time Networks and Object Orientation. The comparison is followed by an illustration of two different design solutions of an existing real-time distributed surveillance system called ADVISOR. One of the design solutions, based on Object Oriented concepts, uses CORBA as a means for the integration and distribution characteristics of the system. The other design solution, based on Real-Time Networks, uses DORIS methodology as a solution for the design of the system. Once the justification over the selection is done, a' novel design of a generic visual surveillance system using the proposed Real-Time Networks method is presented. Finally, the conclusions and future work are explained in the last chapter.
6
Siddiqui, Abdul Jabbar. "A Robust Vehicle Make and Model Recognition System for ITS Applications." Thesis, Université d'Ottawa / University of Ottawa, 2015. http://hdl.handle.net/10393/33124.
Full textAbstract:
A real-time Vehicle Make and Model Recognition (VMMR) system is a significant component of security applications in Intelligent Transportation Systems (ITS). A highly accurate and real-time VMMR system significantly reduces the overhead cost of resources otherwise required. In this thesis, we present a VMMR system that provides very high classification rates and is robust to challenges like low illumination, occlusions, partial and non-frontal views. These challenges are encountered in realistic environments and high security areas like parking lots and public spaces (e.g., malls, stadiums, and airports). The VMMR problem is a multi-class classification problem with a peculiar set of issues and challenges like multiplicity, inter- and intra-make ambiguity among various vehicles makes and models, which need to be solved in an efficient and reliable manner to achieve a highly robust VMMR system. To reliably overcome the ambiguity challenges, a global features representation approach based on the Bag-of-Features paradigm is proposed. We extract key features from different make-model classes in an optimized dictionary, through two different dictionary building strategies. We represent different samples from each class with respect to the learned dictionary. We also present two classification schemes based on multi-class Support Vector Machines (SVMs): (1) Single multi-class SVM and (2) Attribute Bagging-based Ensemble of multi-class SVMs. These classification schemes allow simultaneous learning of the differences between global representations of different classes and the similarities between different shapes or generations within a same make-model class, to further overcome the multiplicity challenges for real-time application.
Extensive experiments conducted using our approaches yield superior results for images that were occluded, under low illumination, partial camera views, or even non-frontal views, available in a recently published real-world VMMR dataset. The approaches presented herewith provide a highly accurate VMMR system for real-time applications in realistic environments.
7
Liu, Junbin. "Distributed low-power image processing in wireless sensor networks for intelligent video surveillance applications." Thesis, Queensland University of Technology, 2012. https://eprints.qut.edu.au/63311/1/Junbin_Liu_Thesis.pdf.
Full textAbstract:
Distributed Wireless Smart Camera (DWSC) network is a special type of Wireless Sensor Network (WSN) that processes captured images in a distributed manner. While image processing on DWSCs sees a great potential for growth, with its applications possessing a vast practical application domain such as security surveillance and health care, it suffers from tremendous constraints. In addition to the limitations of conventional WSNs, image processing on DWSCs requires more computational power, bandwidth and energy that presents significant challenges for large scale deployments. This dissertation has developed a number of algorithms that are highly scalable, portable, energy efficient and performance efficient, with considerations of practical constraints imposed by the hardware and the nature of WSN. More specifically, these algorithms tackle the problems of multi-object tracking and localisation in distributed wireless smart camera net- works and optimal camera configuration determination.
Addressing the first problem of multi-object tracking and localisation requires solving a large array of sub-problems. The sub-problems that are discussed in this dissertation are calibration of internal parameters, multi-camera calibration for localisation and object handover for tracking. These topics have been covered extensively in computer vision literatures, however new algorithms must be invented to accommodate the various constraints introduced and required by the DWSC platform.
A technique has been developed for the automatic calibration of low-cost cameras which are assumed to be restricted in their freedom of movement to either pan or tilt movements. Camera internal parameters, including focal length, principal point, lens distortion parameter and the angle and axis of rotation, can be recovered from a minimum set of two images of the camera, provided that the axis of rotation between the two images goes through the camera's optical centre and is parallel to either the vertical (panning) or horizontal (tilting) axis of the image.
For object localisation, a novel approach has been developed for the calibration of a network of non-overlapping DWSCs in terms of their ground plane homographies, which can then be used for localising objects. In the proposed approach, a robot travels through the camera network while updating its position in a global coordinate frame, which it broadcasts to the cameras. The cameras use this, along with the image plane location of the robot, to compute a mapping from their image planes to the global coordinate frame. This is combined with an occupancy map generated by the robot during the mapping process to localised objects moving within the network.
In addition, to deal with the problem of object handover between DWSCs of non-overlapping fields of view, a highly-scalable, distributed protocol has been designed. Cameras that follow the proposed protocol transmit object descriptions to a selected set of neighbours that are determined using a predictive forwarding strategy. The received descriptions are then matched at the subsequent camera on the object's path using a probability maximisation process with locally generated descriptions.
The second problem of camera placement emerges naturally when these pervasive devices are put into real use. The locations, orientations, lens types etc. of the cameras must be chosen in a way that the utility of the network is maximised (e.g. maximum coverage) while user requirements are met. To deal with this, a statistical formulation of the problem of determining optimal camera configurations has been introduced and a Trans-Dimensional Simulated Annealing (TDSA) algorithm has been proposed to effectively solve the problem.
8
Lorio, Berino. "Towards a non-intrusive traffic surveillance system using digital image processing." Thesis, Stellenbosch : Stellenbosch University, 2001. http://hdl.handle.net/10019.1/52589.
Full textAbstract:
Thesis (MScEng)--Stellenbosch University, 2001.ENGLISH ABSTRACT: With the increased focus on the use of innovative and state-of-the-art technology in Intelligent Transport Systems (ITS), the need for more accurate and more detailed road traffic flow data has become apparent. Data obtained from vehicle detector loops, which merely act as vehicle presence sensors, is neither reliable nor accurate enough anymore. This type of sensor poses the problem that it has to be inserted into the road surface; temporarily obstructing traffic flows, and has to be replaced after pavement reconstruction. One of the solutions to this problem is to develop a traffic surveillance system that uses video image processing. In cities where Intelligent Transport Systems are used extensively, roadways are monitored through Closed Circuit Television Cameras (CCTV) that are closely watched by traffic control centre personnel. These cameras are mounted on posts on the roadside. These cameras can serve a dual purpose, being used for both human monitoring and as inputs to Video Image Processing Systems. In this study some of the digital image processing techniques that could be used in a traffic surveillance system were investigated. This report leads the reader through the various steps in the processing of a scene by a traffic surveillance system based on feature tracking, and discusses the pitfalls and problems that are experienced. The tracker was tested using three image sequences and the results are presented in the final chapter of this report.
AFRIKAANSE OPSOMMING: Met die toenemende fokus op die gebruik van innoverende oplossings en gevorderde tegnologie in Intelligente Vervoerstelsels, het die noodsaaklikheid van akkurater en meer gedetailleerde padverkeer vloeidata duidelik geword. Data wat verkry word d.m.v. voertuig deteksie lusse, wat alleenlik voertuig teenwoordigheid/afwesigheid meet, is nie meer akkuraat of betroubaar genoeg nie. Hierdie tipe sensors het egter die nadeel dat dit in die plaveisel ingesny moet word, dus vloei tydelik kan belemmer, en moet vervang word elke keer as plaveisel rekonstruksie gedoen word. Een van die oplossings vir hierdie probleem is om 'n verkeers waarnemingstelsel te ontwikkel wat van videobeeldverwerking gebruik maak. In stede waar van uitgebreide intelligente verkeerstelsels gebruik gemaak word, word paaie gemonitor d.m.v. geslote baan televisiekameras wat op pale langs die paaie aangebring is. Personeellede van die verkeers beheer sentrum hou dan die inkomende televisiebeelde dop. Hierdie kameras kan 'n dubelle rol vervul deurdat dit vir beide menslike waarneming en as invoer in 'n video-beeldverwerking stelsel gebruik kan word. In hierdie studie was verskeie digitale beeldverwerking tegnieke wat gebruik kan word in 'n verkeers waarnemingstelsel ondersoek. Hierdie verslag lei die leser deur die verskeie stappe in die verwerking van 'n toneel deur 'n verkeers waarneming stelsel wat gebaseer is op die volg van kenmerke. Die verslag beskryf ook die slaggate en probleme wat ondervind word. Die voertuig volger was getoets deur van drie reekse beelde gebruik te maak en die resultate word weergegee in die finale hoodfstuk van hierdie verslag.
9
Sugianto, Nehemia. "Responsible AI for Automated Analysis of Integrated Video Surveillance in Public Spaces." Thesis, Griffith University, 2021. http://hdl.handle.net/10072/409586.
Full textAbstract:
Understanding customer experience in real-time can potentially support people’s safety and comfort while in public spaces. Existing techniques, such as surveys and interviews, can only analyse data at specific times. Therefore, organisations that manage public spaces, such as local government or business entities, cannot respond immediately when urgent actions are needed. Manual monitoring through surveillance cameras can enable organisation personnel to observe people. However, fatigue and human distraction during constant observation cannot ensure reliable and timely analysis. Artificial intelligence (AI) can automate people observation and analyse their movement and any related properties in real-time. Analysing people’s facial expressions can provide insight into how comfortable they are in a certain area, while analysing crowd density can inform us of the area’s safety level. By observing the long-term patterns of crowd density, movement, and spatial data, the organisation can also gain insight to develop better strategies for improving people’s safety and comfort.
There are three challenges to making an AI-enabled video surveillance system work well in public spaces. First is the readiness of AI models to be deployed in public space settings. Existing AI models are designed to work in generic/particular settings and will suffer performance degradation when deployed in a real-world setting. Therefore, the models require further development to tailor them for the specific environment of the targeted deployment setting. Second is the inclusion of AI continual learning capability to adapt the models to the environment. AI continual learning aims to learn from new data collected from cameras to adapt the models to constant visual changes introduced in the setting. Existing continuous learning approaches require long-term data retention and past data, which then raise data privacy issues. Third, most of the existing AI-enabled surveillance systems rely on centralised processing, meaning data are transmitted to a central/cloud machine for video analysis purposes. Such an approach involves data privacy and security risks. Serious data threats, such as data theft, eavesdropping or cyberattack, can potentially occur during data transmission.
This study aims to develop an AI-enabled intelligent video surveillance system based on deep learning techniques for public spaces established on responsible AI principles. This study formulates three responsible AI criteria, which become the guidelines to design, develop, and evaluate the system. Based on the criteria, a framework is constructed to scale up the system over time to be readily deployed in a specific real-world environment while respecting people’s privacy. The framework incorporates three AI learning approaches to iteratively refine the AI models within the ethical use of data. First is the AI knowledge transfer approach to adapt existing AI models from generic deployment to specific real-world deployment with limited surveillance datasets. Second is the AI continuous learning approach to continuously adapt AI models to visual changes introduced by the environment without long-period data retention and the need for past data. Third is the AI federated learning approach to limit sensitive and identifiable data transmission by performing computation locally on edge devices rather than transmitting to the central machine.
This thesis contributes to the study of responsible AI specifically in the video surveillance context from both technical and non-technical perspectives. It uses three use cases at an international airport as the application context to understand passenger experience in real-time to ensure people’s safety and comfort. A new video surveillance system is developed based on the framework to provide automated people observation in the application context. Based on real deployment using the airport’s selected cameras, the evaluation demonstrates that the system can provide real-time automated video analysis for three use cases while respecting people’s privacy. Based on comprehensive experiments, AI knowledge transfer can be an effective way to address limited surveillance datasets issue by transferring knowledge from similar datasets rather than training from scratch on surveillance datasets. It can be further improved by incrementally transferring knowledge from multi-datasets with smaller gaps rather than a one-stage process. Learning without Forgetting is a viable approach for AI continuous learning in the video surveillance context. It consistently outperforms fine-tuning and joint-training approaches with lower data retention and without the need for past data. AI federated learning can be a feasible solution to allow continuous learning in the video surveillance context without compromising model accuracy. It can obtain comparable accuracy with quicker training time compared to joint-training.Thesis (PhD Doctorate)
Doctor of Philosophy (PhD)
Dept Bus Strategy & Innovation
Griffith Business School
Full Text
10
Chen, Jiandan. "An Intelligent Multi Sensor System for a Human Activities Space---Aspects of Quality Measurement and Sensor Arrangement." Doctoral thesis, Karlskrona : Blekinge Institute of Technology, 2011. http://urn.kb.se/resolve?urn=urn:nbn:se:bth-00487.
Full textAbstract:
In our society with its aging population, the design and implementation of a highperformance distributed multi-sensor and information system for autonomous physical services become more and more important. In line with this, this thesis proposes an Intelligent Multi-Sensor System, IMSS, that surveys a human activities space to detect and identify a target for a specific service. The subject of this thesis covers three main aspects related to the set-up of an IMSS: an improved depth measurement and reconstruction method and its related uncertainty, a surveillance and tracking algorithm and finally a way to validate and evaluate the proposed methods and algorithms. The thesis discusses how a model of the depth spatial quantisation uncertainty can be implemented to optimize the configuration of a sensor system to capture information of the target objects and their environment with required specifications. The thesis introduces the dithering algorithm which significantly reduces the depth reconstruction uncertainty. Furthermore, the dithering algorithm is implemented on a sensor-shifted stereo camera, thus simplifying depth reconstruction without compromising the common stereo field of view. To track multiple targets continuously, the Gaussian Mixture Probability Hypothesis Density, GM-PHD, algorithm is implemented with the help of vision and Radio Frequency Identification, RFID, technologies. The performance of the tracking algorithm in a vision system is evaluated by a circular motion test signal. The thesis introduces constraints to the target space, the stereo pair characteristics and the depth reconstruction accuracy to optimize the vision system and to control the performance of surveillance and 3D reconstruction through integer linear programming. The human being within the activity space is modelled as a tetrahedron, and a field of view in spherical coordinates are used in the control algorithms. In order to integrate human behaviour and perception into a technical system, the proposed adaptive measurement method makes use of the Fuzzily Defined Variable, FDV. The FDV approach enables an estimation of the quality index based on qualitative and quantitative factors for image quality evaluation using a neural network. The thesis consists of two parts, where Part I gives an overview of the applied theory and research methods used, and Part II comprises the eight papers included in the thesis.
11
Saggese, Alessia. "Detecting and indexing moving objects for Behavior Analysis by Video and Audio Interpretation." Doctoral thesis, Universita degli studi di Salerno, 2014. http://hdl.handle.net/10556/1487.
Full textAbstract:
2012 - 2013In the last decades we have assisted to a growing need for security in many public environments. According to a study recently conducted by the European Security Observatory, one half of the entire population is worried about the crime and requires the law enforcement to be protected. This consideration has lead the proliferation of cameras and microphones, which represent a suitable solution for their relative low cost of maintenance, the possibility of installing them virtually everywhere and, finally, the capability of analysing more complex events. However, the main limitation of this traditional audiovideo surveillance systems lies in the so called psychological overcharge issue of the human operators responsible for security, that causes a decrease in their capabilities to analyse raw data flows from multiple sources of multimedia information; indeed, as stated by a study conducted by Security Solutions magazine, after 12 minutes of continuous video monitoring, a guard will often miss up to 45% of screen activity. After 22 minutes of video, up to 95% is overlooked. For the above mentioned reasons, it would be really useful to have available an intelligent surveillance system, able to provide images and video with a semantic interpretation, for trying to bridge the gap between their low-level representation in terms of pixels, and the high-level, natural language description that a human would give about them. On the other hand, this kind of systems, able to automatically understand the events occurring in a scene, would be really useful in other application fields, mainly oriented to marketing purposes. Especially in the last years, a lot of business intelligent applications have been installed for assisting decision makers and for giving an organization’s employees, partners and suppliers easy access to the information they need to effectively do their jobs... [edited by author]
XII n.s.
12
Magaia, Lourenco Lazaro. "A video-based traffic monitoring system." Thesis, Stellenbosch : University of Stellenbosch, 2006. http://hdl.handle.net/10019.1/1243.
Full textAbstract:
Thesis (PhD (Mathematical Sciences. Applied Mathematics))--University of Stellenbosch, 2006.This thesis addresses the problem of bulding a video-based traffic monitoring system. We employ clustering, trackiing and three-dimensional reconstruction of moving objects over a long image sequence. We present an algorithms that robustly recovers the motion and reconstructs three-dimensional shapes from a sequence of video images, Magaia et al [91]. The problem ...
13
Sutor, S. R. (Stephan R. ). "Large-scale high-performance video surveillance." Doctoral thesis, Oulun yliopisto, 2014. http://urn.fi/urn:isbn:9789526205618.
Full textAbstract:
Abstract The last decade was marked by a set of harmful events ranging from economical crises to organized crime, acts of terror and natural catastrophes. This has led to a paradigm transformation concerning security. Millions of surveillance cameras have been deployed, which led to new challenges, as the systems and operations behind those cameras could not cope with the rapid growth in number of video cameras and systems. Looking at today’s control rooms, often hundreds or even thousands of cameras are displayed, overloading security officers with irrelevant information. The purpose of this research was the creation of a novel video surveillance system with automated analysis mechanisms which enable security authorities and their operators to cope with this information flood. By automating the process, video surveillance was transformed into a proactive information system. The progress in technology as well as the ever increasing demand in security have proven to be an enormous driver for security technology research, such as this study. This work shall contribute to the protection of our personal freedom, our lives, our property and our society by aiding the prevention of crime and terrorist attacks that diminish our personal freedom. In this study, design science research methodology was utilized in order to ensure scientific rigor while constructing and evaluating artifacts. The requirements for this research were sought in close cooperation with high-level security authorities and prior research was studied in detail. The created construct, the “Intelligent Video Surveillance System”, is a distributed, highly-scalable software framework, that can function as a basis for any kind of high-performance video surveillance system, from installations focusing on high-availability to flexible cloud-based installation that scale across multiple locations and tens of thousands of cameras. First, in order to provide a strong foundation, a modular, distributed system architecture was created, which was then augmented by a multi-sensor analysis process. Thus, the analysis of data from multiple sources, combining video and other sensors in order to automatically detect critical events, was enabled. Further, an intelligent mobile client, the video surveillance local control, which addressed remote access applications, was created. Finally, a wireless self-contained surveillance system was introduced, a novel smart camera concept that enabled ad hoc and mobile surveillance. The value of the created artifacts was proven by evaluation at two real-world sites: An international airport, which has a large-scale installation with high-security requirements, and a security service provider, offering a multitude of video-based services by operating a video control center with thousands of cameras connectedTiivistelmä Viime vuosikymmen tunnetaan vahingollisista tapahtumista alkaen talouskriiseistä ja ulottuen järjestelmälliseen rikollisuuteen, terrori-iskuihin ja luonnonkatastrofeihin. Tämä tilanne on muuttanut suhtautumista turvallisuuteen. Miljoonia valvontakameroita on otettu käyttöön, mikä on johtanut uusiin haasteisiin, koska kameroihin liittyvät järjestelmät ja toiminnot eivät pysty toimimaan yhdessä lukuisien uusien videokameroiden ja järjestelmien kanssa. Nykyajan valvontahuoneissa voidaan nähdä satojen tai tuhansien kameroiden tuottavan kuvaa ja samalla runsaasti tarpeetonta informaatiota turvallisuusvirkailijoiden katsottavaksi. Tämän tutkimuksen tarkoitus oli luoda uusi videovalvontajärjestelmä, jossa on automaattiset analyysimekanismit, jotka mahdollistavat turva-alan toimijoiden ja niiden operaattoreiden suoriutuvan informaatiotulvasta. Automaattisen videovalvontaprosessin avulla videovalvonta muokattiin proaktiiviseksi tietojärjestelmäksi. Teknologian kehitys ja kasvanut turvallisuusvaatimus osoittautuivat olevan merkittävä ajuri turvallisuusteknologian tutkimukselle, kuten tämä tutkimus oli. Tämä tutkimus hyödyttää yksittäisen ihmisen henkilökohtaista vapautta, elämää ja omaisuutta sekä yhteisöä estämällä rikoksia ja terroristihyökkäyksiä. Tässä tutkimuksessa suunnittelutiedettä sovellettiin varmistamaan tieteellinen kurinalaisuus, kun artefakteja luotiin ja arvioitiin. Tutkimuksen vaatimukset perustuivat läheiseen yhteistyöhön korkeatasoisten turva-alan viranomaisten kanssa, ja lisäksi aiempi tutkimus analysoitiin yksityiskohtaisesti. Luotu artefakti - ’älykäs videovalvontajärjestelmä’ - on hajautettu, skaalautuva ohjelmistoviitekehys, joka voi toimia perustana monenlaiselle huipputehokkaalle videovalvontajärjestelmälle alkaen toteutuksista, jotka keskittyvät saatavuuteen, ja päättyen joustaviin pilviperustaisiin toteutuksiin, jotka skaalautuvat useisiin sijainteihin ja kymmeniin tuhansiin kameroihin. Järjestelmän tukevaksi perustaksi luotiin hajautettu järjestelmäarkkitehtuuri, jota laajennettiin monisensorianalyysiprosessilla. Siten mahdollistettiin monista lähteistä peräisin olevan datan analysointi, videokuvan ja muiden sensorien datan yhdistäminen ja automaattinen kriittisten tapahtumien tunnistaminen. Lisäksi tässä työssä luotiin älykäs kännykkäsovellus, videovalvonnan paikallinen kontrolloija, joka ohjaa sovelluksen etäkäyttöä. Viimeksi tuotettiin langaton itsenäinen valvontajärjestelmä – uudenlainen älykäs kamerakonsepti – joka mahdollistaa ad hoc -tyyppisen ja mobiilin valvonnan. Luotujen artefaktien arvo voitiin todentaa arvioimalla ne kahdessa reaalimaailman ympäristössä: kansainvälinen lentokenttä, jonka laajamittaisessa toteutuksessa on korkeat turvavaatimukset, ja turvallisuuspalveluntuottaja, joka tarjoaa moninaisia videopohjaisia palveluja videovalvontakeskuksen avulla käyttäen tuhansia kameroita
14
Saggese, Alessia. "Detecting and indexing moving objects for behavior analysis by video and audio interpretation." Doctoral thesis, Caen, 2014. http://www.theses.fr/2014CAEN2021.
Full textAbstract:
Dans les dernières décennies, nous avons été témoin d'un besoin grandissant de sécurité dans les espaces publics. La limitation principale induite par les systèmes de vidéo surveillance réside dans la surcharge cognitive des opérateurs humains chargés de la sécurité, ce qui diminue leur capacités à analyser le flux d'information émanant de sources multimédia multiples. Pour ces raisons, nous proposons dans cette thèse un système de surveillance intelligent capable d'associer des images et des vidéos à une interprétation sémantique afin de faire le lien entre des représentations bas niveau, sous forme de pixels, et le haut niveau correspondant à une description en langage naturel qu'un être humain pourrait faire d'une scène. Plus précisément, les travaux proposés débutent par l'analyse des vidéos et par l'extraction des trajectoires des objets présents dans la scène. Une fois extraites, ce grand nombre de trajectoires doit être indexé et stocké afin d'augmenter la performance du système durant la phase de reconnaissance. En outre, l'opérateur humain est informé immédiatement si un comportement anormal est observé. Tandis que l'information extraite des vidéos n'est pas suffisante ou n'est pas suffisamment fiable, le système proposé est enrichi par un module en charge de la reconnaissance des événements sonores tels que des tirs, des cris ou des vitres cassées. Chaque module proposé a été à la fois testé sur des jeux de données standards mais aussi dans un environnement réel ; les résultats obtenus, tout comme l'application des méthodes proposées dans un contexte réel, permettent de confirmer la contribution de nos travaux à l'état de l'artIn the last decades we have assisted to a growing need for security in many public environments. The main limitation of this traditional audio-video surveillance systems lies in the so called psychological overcharge issue of the human operators responsible for security, that causes a decrease in their capabilities to analyse raw data flows from multiple sources of multimedia information. For the above mentioned reasons, in this thesis we propose an intelligent surveillance system able to provide images and video with a semantic interpretation, for trying to bridge the gap between their low-level representation in terms of pixels, and the high-level, natural language description that a human would give about them. In particular, the proposed framework starts by analysing the videos and by extracting the trajectories of the objects populating the scene. Once extracted, this large amount of trajectories needs to be indexed and properly stored in order to improve the overall performance of the system during the retrieving. Furthermore, the human operator is informed as soon as an abnormal behaviour occurs. Whereas the information extracted from the videos are not sufficient or not sufficiently reliable, the proposed system in enriched by a module in charge of recognizing audio events, such as shoots, screams or broken glasses. Each proposed module has been tested both over standard datasets and in real environments; the promising obtained results confirm the advance with respect to the state of the art, as well as the applicability of the proposed method in real scenarios
15
Asif, Muhammad. "Video analytics for intelligent surveillance systems." Thesis, University of Strathclyde, 2010. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.530322.
Full text
16
Aasen, Thomas Aron. "Case Based Surveillance System." Thesis, Norges teknisk-naturvitenskapelige universitet, Institutt for datateknikk og informasjonsvitenskap, 2006. http://urn.kb.se/resolve?urn=urn:nbn:no:ntnu:diva-15843.
Full textAbstract:
Many problems in the field of automatic video surveillance exists today. Some have yet to be overcome. One of these problems is how a computer system automatically can determine if a situation should cause an alarm or not. To resolve this problem, the use of Case-based reasoning (CBR) is proposed. CBR is a technique that allows a system to reason about different situations and to learn from them. The aim is to produce a system that utilizes these abilities. The system should learn to recognize the situations that causes different alarms. When a situation is recognized and categorized, these false alarms can be completely avoided. This master thesis explains and shows the advantages of using such a system together with advanced image processing techniques.
17
LAVI, SEFIDGARI BAHRAM. "Person Re-Identification Techniques for Intelligent Video Surveillance Systems." Doctoral thesis, Università degli Studi di Cagliari, 2018. http://hdl.handle.net/11584/255952.
Full textAbstract:
Nowadays, intelligent video-surveillance is one of the most active research fields in com- puter vision and machine learning techniques which provides useful tools for surveillance operators and forensic video investigators. Person re-identification is among these tools; it consists of recognizing whether an individual has already been observed over a network of cameras. This tool can also be employed in various possible applications, e.g., off-line retrieval of all the video-sequences showing an individual of interest whose image is given as query, or on-line pedestrian tracking over multiple cameras. For the off-line retrieval applications, one of the goals of person re-identification systems is to support video surveillance operators and forensic investigators to find an individual of interest in videos acquired by a network of non-overlapping cameras. This is attained by sorting images of previously ob- served individuals for decreasing values of their similarity with a given probe individual.
This task is typically achieved by exploiting the clothing appearance, in which a classical biometric methods like the face recognition is impeded to be practical in real-world video surveillance scenarios, because of low-quality of acquired images. Existing clothing appearance descriptors, together with their similarity measures, are mostly aimed at im- proving ranking quality. These methods usually are employed as part-based body model in order to extract image signature that might be independently treated in different body parts (e.g. torso and legs). Whereas, it is a must that a re-identification model to be robust and discriminate on individual of interest recognition, the issue of the processing time might also be crucial in terms of tackling this task in real-world scenarios. This issue can be also seen from two different point of views such as processing time to construct a model (aka descriptor generation); which usually can be done off-line, and processing time to find the correct individual from bunch of acquired video frames (aka descriptor matching); which is the real-time procedure of the re-identification systems.
This thesis addresses the issue of processing time for descriptor matching, instead of im- proving ranking quality, which is also relevant in practical applications involving interaction with human operators. It will be shown how a trade-off between processing time and rank- ing quality, for any given descriptor, can be achieved through a multi-stage ranking approach inspired by multi-stage approaches to classification problems presented in pattern recogni- tion area, which it is further adapting to the re-identification task as a ranking problem. A discussion of design criteria is therefore presented as so-called multi-stage re-identification systems, and evaluation of the proposed approach carry out on three benchmark data sets, using four state-of-the-art descriptors. Additionally, by concerning to the issue of processing time, typical dimensional reduction methods are studied in terms of reducing the processing time of a descriptor where a high-dimensional feature space is generated by a specific person re-identification descriptor. An empirically experimental result is also presented in this case, and three well-known feature reduction methods are applied them on two state-of-the-art descriptors on two benchmark data sets.
18
DELUSSU, RITA. "Human Centered Computer Vision Techniques for Intelligent Video Surveillance Systems." Doctoral thesis, Università degli Studi di Cagliari, 2021. http://hdl.handle.net/11584/309042.
Full textAbstract:
Nowadays, intelligent video surveillance systems are being developed to support human operators in different monitoring and investigation tasks. Although relevant results have been achieved by the research community in several computer vision tasks, some real applications still exhibit several open issues. In this context, this thesis focused on two challenging computer vision tasks: person re-identification and crowd counting.
Person re-identification aims to retrieve images of a person of interest, selected by the user, in different locations over time, reducing the time required to the user to analyse all the available videos. Crowd counting consists of estimating the number of people in a given image or video. Both tasks present several complex issues. In this thesis, a challenging video surveillance application scenario is considered in which it is not possible to collect and manually annotate images of a target scene (e.g., when a new camera installation is made by Law Enforcement Agency) to train a supervised model. Two human centered solutions for the above mentioned tasks are then proposed, in which the role of the human operators is fundamental.
For person re-identification, the human-in-the-loop approach is proposed, which exploits the operator feedback on retrieved pedestrian images during system operation, to improve system's effectiveness. The proposed solution is based on revisiting relevance feedback algorithms for content-based image retrieval, and on developing a specific feedback protocol, to find a trade-off between the human effort and re-identification performance.
For crowd counting, the use of a synthetic training set is proposed to develop a scene-specific model, based on a minimal amount of information of the target scene required to the user. Both solutions are empirically investigated using state-of-the-art supervised models based on Convolutional Neural Network, on benchmark data sets.
19
Diaz, Solis David Alejandro. "Financial market monitoring and surveillance systems framework : a service systems and business intelligence approach." Thesis, University of Manchester, 2012. https://www.research.manchester.ac.uk/portal/en/theses/financial-market-monitoring-and-surveillance-systems-frameworka-service-systems-and-business-intelligence-approach(47e568f8-3024-4ca3-8114-5d183be3edb8).html.
Full textAbstract:
The thesis introduces a framework for analysing market monitoring and surveillance systems in order to provide a common foundation for researchers and practitioners to specify, design, implement, compare and evaluate such systems. The proposed framework serves as a reference map for researchers and practitioners to position their work in the context of market monitoring and surveillance, resulting in a useful instrument for the analysis, testing and management of such systems. More specifically, the thesis examines the new requirements for the operation of financial markets, the role of technologies, the recent consultations on the structure and governance of EU and US markets, as well as, future usage scenarios and emerging technologies. It examines the context in which market monitoring and market surveillance systems are currently been used. It reports on their processes, performance, and on the organisational and regulatory environments in which they exist. Furthermore, it develops a set of taxonomies which cover the majority of the concepts of market manipulation, market monitoring, market surveillance, entities, technologies and actors that are relevant for the work in this thesis. Building on the gaps and limitations of the current systems, it proposes a new framework following the Design Science methodology. The usefulness of the framework is evaluated through four critical case studies, which not only help to understand with practical exercises the way how markets monitoring and surveillance systems work, but also to investigate their weaknesses, potential evolution and ways to improve them. For each case study, the thesis develops a fully working prototype tested using a sample prosecution case and evaluated in terms of the appropriateness and suitability of the proposed framework. Finally, implications relating to policies, procedures and future market structures are discussed followed by suggestions for future research.
20
Jönsson, Jonatan, and Felix Stenbäck. "Fence surveillance with convolutional neural networks." Thesis, Högskolan i Halmstad, Akademin för informationsteknologi, 2018. http://urn.kb.se/resolve?urn=urn:nbn:se:hh:diva-37116.
Full textAbstract:
Broken fences is a big security risk for any facility or area with strict security standards. In this report we suggest a machine learning approach to automate the surveillance for chain-linked fences. The main challenge is to classify broken and non-broken fences with the help of a convolution neural network. Gathering data for this task is done by hand and the dataset is about 127 videos at 26 minutes length total on 23 different locations. The model and dataset are tested on three performances traits, scaling, augmentation improvement and false rate. In these tests we concluded that nearest neighbor increased accuracy. Classifying with fences that has been included in the training data a false rate that was low, about 1%. Classifying with fences that are unknown to the model produced a false rate of about 90%. With these results we concludes that this method and dataset is useful under the right circumstances but not in an unknown environment.
21
Laarabi, Mohamed Haitam. "Optimisation multicritère des itinéraires pour transport des marchandises dangereuses en employant une évaluation en logique floue du risque et la simulation du trafic à base d'agents." Thesis, Paris, ENMP, 2014. http://www.theses.fr/2014ENMP0074/document.
Full textAbstract:
Chaque jour des milliers de camions transportant des centaines de milliers de tonnes de marchandises dangereuses par diverses modalités. Toutefois, le terme “dangereux” indique une adversité intrinsèque qui caractérise ces produits transportés, et qui peuvent se manifester lors d'un accident entraînant la fuite d'une substance dangereuse. Dans une telle situation, les conséquences peuvent nuire à l'environnement et létal pour l'humain.L'importance des marchandises dangereuses revient aux bénéfices économiques considérables générés. En fait, on ne peut nier la contribution du transport des produits dérivés de combustibles fossiles, ce qui représente plus de 60% des marchandises dangereuses transportées en Europe. Eni, la société italienne leader de pétrochimie, gère chaque jour une flotte d'environ 1.500 camions, qui effectuent de nombreuses expéditions. Pourtant la distribution de produits pétroliers est une activité à grande risques, et tout accident lors du transport peut entraîner de graves conséquences.Consciente des enjeux, la division Eni R&M - Logistique Secondaire, historiquement actif au siège de Gênes, collabore depuis 2002 avec le DIBRIS à l'Université de Gênes, et le CRC à Mines ParisTech, dans le but d'étudier les améliorations possibles en matière de sûreté dans le transport de marchandises dangereuses. Au fil des ans, cette collaboration a permis le développement d'un système d'information et décisionnel. Le composant principal de ce système est une plate-forme de surveillance de la flotte Eni appelé TIP (Transport Integrated Platform), pour livrer les produits vers les points de distributions. Ces véhicules sont équipés d'un dispositif capable de transmettre des flux de données en temps réel en utilisant un modem GPRS. Les données transmises peuvent être de nature différente et contenir des informations sur l'état du véhicule, le produit et les événements détectés durant l'expédition. Ces données sont destinées à être reçues par des serveurs centralisés puis traitées et stockées, afin de soutenir diverses applications du TIP.Dans ce contexte, les études menées tout au long de la thèse sont dirigés vers le développement d'une proposition visant à réduire davantage les risques liés au transport de marchandises dangereuses. En d'autres termes, un modèle basé sur le compromis entre les facteurs économiques et sûretés pour le choix de l'itinéraire. L'objectif est motivé par la nécessité de soutenir les règlements et les normes de sécurité existantes, car ils ne garantissent pas totalement contre les accidents entrainant des marchandises dangereuses.L'objectif est effectué en prenant en compte le système existant comme base pour l'élaboration d'un système de transport intelligent (STI) regroupant plusieurs plates-formes logicielles. Ces plates-formes doivent permettre aux planificateurs et aux décideurs de suivre en temps réel leur flotte, à évaluer les risques et tous les itinéraires possibles, de simuler et de créer différents scénarios, et d'aider à trouver des solutions à des problèmes particuliers.Tout au long de cette thèse, je souligne la motivation pour ce travail de recherche, les problématiques, et les défis de transport de marchandises dangereuses. Je présente le TIP comme le noyau de l'architecture proposée du STI. Pour les besoins de la simulation, les véhicules virtuels sont injectés dans le système. La gestion de la collecte des données a été l'objet d'une amélioration technique pour plus de fiabilité, d'efficacité et d'évolutivité dans le cadre de la surveillance en temps réel. Enfin, je présente une explication systématique de la méthode d'optimisation des itinéraires considérant les critères économiques et de risques. Le risque est évalué en fonction de divers facteurs notamment la fréquence d'accidents entrainant des marchandises dangereuses, et ses conséquences. La quantification de l'incertitude dans l'évaluation des risques est modélisée en utilisant la théorie des ensembles flousEveryday thousands of trucks transporting hundreds of thousands of tons of dangerous goods by various modalities and both within and across nations. However, the term “dangerous” indicates an intrinsic adversity that characterize these products, which can manifest in an accident leading to release of a hazardous substance (e.g. radioactive, flammable, explosive etc.). In this situation, the consequences can be lethal to human beings, other living organisms and damage the environment and public/private properties.The importance of dangerous goods boils down to the significant economic benefits that generates. In fact, one cannot deny the contribution of the transport of all fossil fuel derived product, which represents more than 60% of dangerous goods transported in Europe. Eni, the Italian leading petrochemical company, every day operates a fleet of about 1,500 trucks, which performs numerous trips from loading terminals to filling stations. Distribution of petroleum products is a risky activity, and an accident during the transportation may lead to serious consequences.Aware of what is at stake, the division Eni R&M - Logistics Secondary, historically active in Genoa headquarters, is collaborating since 2002 with the DIBRIS department at University of Genoa, and the CRC at Mines ParisTech, with the purpose of studying possible improvements regarding safety in transport of dangerous goods, particularly petroleum products. Over years, this collaboration has led to the development of different technologies and mainly to an information and decision support system. The major component of this system is a platform for monitoring Eni fleet, at the national level, to deliver the products to the distribution points, called the Transport Integrated Platform (TIP). These vehicles are equipped with a device capable of transmitting data stream in real-time using a GPRS modem. The data transmitted can be of different nature and contain information about the state of the vehicle and occurred events during the trip. These data are intended to be received by centralized servers then get processed and stored, in order to support various applications within the TIP.With this in mind, the studies undertaken throughout the thesis are directed towards the development of a proposal to further minimize the risk related to the transportation of dangerous goods. In other words, a trade-off based model for route selection taking into consideration economic and safety factors. The objective is prompted by the need to support existent regulations and safety standards, which does not assure a full warranty against accidents involving dangerous goods.The goal is carried out by considering the existent system as basis for developing an Intelligent Transportation System (ITS) aggregating multiple software platforms. These platforms should allow planners and decision makers to monitor in real-time their fleet, to assess risk and evaluate all possible routes, to simulate and create different scenarios, and to assist at finding solutions to particular problems.Throughout this dissertation, I highlight the motivation for such research work, the related problem statements, and the challenges in dangerous goods transport. I introduce the TIP as the core for the proposed ITS architecture. For simulation purposes, virtual vehicles are injected into the system. The management of the data collection was the subject of technical improvement for more reliability, efficiency and scalability in real-time monitoring of dangerous goods shipment. Finally, I present a systematic explanation of the methodology for route optimization considering both economic and risk criteria. The risk is assessed based on various factors mainly the frequency of accident leading to hazardous substance release and its consequences. Uncertainty quantification in risk assessment is modelled using fuzzy sets theory
22
Soh, Sze Shiang. "Determining Intelligence, Surveillance and Reconnaissance (ISR) system effectiveness, and integration as part of force protection and system survivability." Thesis, Monterey California. Naval Postgraduate School, 2013. http://hdl.handle.net/10945/37721.
Full textAbstract:
Approved for public release; distribution is unlimitedSituation awareness plays a critical role in all battlefields. It monitors activities, and provides essential information about the battle. It is an operational requirement, high in demand, for the forces to fight the battle smartly and accomplishing the objectives set with minimal casualties. Situation awareness enhances survivability of the fighting forces by avoiding adversary detection and acquisition, achieved via the deployment of a variety of sensors that are part of an effective and integrated ISR system network. This thesis analyzes the impact of ISR system effectiveness and integration on unit survivability, in the context of a combined arms unit. The study was approached using the Nearly Orthogonal Latin Hypercube to generate design points for simulation study. Map Aware Non-uniform Automata (MANA) was used to simulate the behavior of the units in the combined arms unit. During simulation, the parameters are varied to create a changing situation picture, as perceived by the troops. This determines the impact on survivability, by measuring the force exchange ratio between the RED and BLUE force, once the simulation is completed. The sensor capabilities and level of integration between the ISR sensors in the combined arms unit are analyzed based on the simulation results.
23
Montenegro, Martinez Davis. "Diakoptics basée en acteurs pour la simulation, la surveillance et la comande des réseaux intelligents." Thesis, Université Grenoble Alpes (ComUE), 2015. http://www.theses.fr/2015GREAT106/document.
Full textAbstract:
La simulation de systèmes d'énergie est un outil important pour la conception, le développement et l'évaluation de nouvelles architectures et des contrôles grille dans le concept de réseau intelligent pour les dernières décennies. Cet outil a évolué pour répondre aux questions proposées par les chercheurs et les ingénieurs dans les applications de l'industrie, et pour offrant des différentes alternatives pour couvrir plusieurs scénarios réalistes.Aujourd'hui, en raison des progrès récents dans le matériel informatique, la Simulation numérique en temps réel (DRTS) est utilisée pour concevoir des systèmes de puissance, afin de soutenir les décisions prises dans les systèmes de gestion de l'énergie automatisés (SME) et de réduire le délai de commercialisation de produits, entre des autres applications.Les simulations de réseaux électriques peuvent être classées dans les catégories suivantes: (1) la simulation analogique (2) hors simulation de ligne (3) de simulation entièrement numérique (4) la simulation rapide (5) Contrôleur Hardware-In-the-Loop (CHIL) et (6) Puissance Hardware-In-the-Loop (PHIL).Les dernière 3 sont axés sur la simulation Real-Time hardware-in-the-Loop (HIL RT-). Ces catégories portent sur les questions liées à Transitoires électromagnétiques (liste EMT), la simulation de phaseurs ou mixte (phaseur et EMT). Comme mentionné ci-dessus, ces progrès sont possibles en raison de l'évolution des architectures informatiques (matériels et logiciels); Cependant, pour le cas particulier de l'analyse des flux de puissance des réseaux de distribution (DS), il y a encore des défis à résoudre.Les architectures informatiques actuelles sont composées de plusieurs noyaux, laissant derrière lui le paradigme de la programmation séquentielle et conduisant les développeurs de systèmes numériques pour examiner des concepts comme le parallélisme, la concurrence et les événements asynchrones. D'autre part, les méthodes pour résoudre le flux de puissance dynamique des systèmes de distribution considérer le système comme un seul bloc; ainsi, ils utilisent une seule base pour l'analyse des flux de puissance, indépendamment de l'existence de plusieurs cœurs disponibles pour améliorer les performances de la simulation.Répartis dans des procédés en phase et de la séquence, ces procédés ont en caractéristiques communes telles que l'examen d'une seule matrice creuse pour décrire les DS et qu'ils peuvent résoudre simultanément une seule fréquence.Ces caractéristiques font dès les méthodes mentionné sont pas appropriées pour le traitement avec multiple noyaux. En conséquence, les architectures informatiques actuelles sont sous-utilisés, et dégrade la performance des simulateurs lors de la manipulation de grandes DS échelle, changer DS topologie et y compris les modèles avancés, entre autres des activités de la vie réelle.Pour relever ces défis Cette thèse propose une approche appelée A-Diakoptics, qui combine la puissance de Diakoptics et le modèle de l'acteur; le but est de faire toute méthode classique d'analyse de flux d'énergie appropriée pour le traitement multithread. En conséquence, la nature et la complexité du système d'alimentation peuvent être modélisées sans affecter le temps de calcul, même si plusieurs parties du système d'alimentation fonctionnent à une fréquence de base différente comme dans le cas de micro-réseaux à courant continu. Par conséquent, l'analyse des flux de charge dynamique de DS peut être effectuée pour couvrir les besoins de simulation différents tels que la simulation hors ligne, simulation rapide, CHIL et PHIL. Cette méthode est une stratégie avancée pour simuler les systèmes de distribution à grande échelle dans des conditions déséquilibrées; couvrant les besoins de base pour la mise en œuvre d'applications de réseaux intelligentsSimulation of power systems is an important tool for designing, developing and assessment of new grid architectures and controls within the smart grid concept for the last decades. This tool has evolved for answering the questions proposed by academic researchers and engineers in industry applications; providing different alternatives for covering several realistic scenarios. Nowadays, due to the recent advances in computing hardware, Digital Real-Time Simulation (DRTS) is used to design power systems, to support decisions made in automated Energy Management Systems (EMS) and to reduce the Time to Market of products, among other applications.Power system simulations can be classified in the following categories: (1) Analog simulation (2) off line simulation (3) Fully digital simulation (4) Fast simulation (5) Controller Hardware-In-the-Loop (CHIL) simulation and (6) Power Hardware-In-the-Loop (PHIL) simulation. The latest 3 are focused on Real-Time Hardware-In-the-Loop (RT-HIL) simulation. These categories cover issues related to Electromagnetic Transients (EMT), phasor simulation or mixed (phasor and EMT). As mentioned above, these advances are possible due to the evolution of computing architectures (hardware and software); however, for the particular case of power flow analysis of Distribution Systems (DS) there are still challenges to be solved.The current computing architectures are composed by several cores, leaving behind the paradigm of the sequential programing and leading the digital system developers to consider concepts such as parallelism, concurrency and asynchronous events. On the other hand, the methods for solving the dynamic power flow of distribution systems consider the system as a single block; thus they only use a single core for power flow analysis, regardless of the existence of multiple cores available for improving the simulation performance.Divided into phase and sequence frame methods, these methods have in common features such as considering a single sparse matrix for describing the DS and that they can solve a single frequency simultaneously. These features make of the mentioned methods non-suitable for multithread processing. As a consequence, current computer architectures are sub-used, affecting simulator's performance when handling large scale DS, changing DS topology and including advanced models, among others real life activities.To address these challenges this thesis proposes an approach called A-Diakoptics, which combines the power of Diakoptics and the Actor model; the aim is to make any conventional power flow analysis method suitable for multithread processing. As a result, the nature and complexity of the power system can be modeled without affecting the computing time, even if several parts of the power system operate at different base frequency as in the case of DC microgrids. Therefore, the dynamic load flow analysis of DS can be performed for covering different simulation needs such as off-line simulation, fast simulation, CHIL and PHIL. This method is an advanced strategy for simulating large-scale distribution systems in unbalanced conditions; covering the basic needs for the implementation of smart grid applications
24
Conte, Donatello. "Detection, tracking, and behaviour analysis of moving people in intelligent video surveillance systems : a graph based approach." Lyon, INSA, 2006. http://theses.insa-lyon.fr/publication/2006ISAL0033/these.pdf.
Full textAbstract:
In this thesis video surveillance system is proposed. For each step of such systems it presents some innovations as regard as the state of the art in such systems. First of all, a new selectively and adaptively background substraction algorithm has been proposed to adapt the system at illumination and scene changes. Furthermore, some heuristics are proposed to solve object detection problems in real environment shadows, noise, etc. Result show that proposed techniques are robust in terms of quality of solution and, besides, they are efficient in terms of processing time. The main object of the thesis concerns the object tracking phase. In the thesis we propose a new algorithm based on a new representation of the objects : the graph pyramids. This presentation allows the resolutions of occlusions also in complex cases. They are preformed on standard datasets and standard indexes to provide objective results. The results show the approch is promisingDans cette thèse, nous proposons un système de vidéo surveillance qui présente des nouveaux algorithmes de détection d’objets et de suivi d’objets, afin de pallier les principaux problèmes qui se présentent dans le développement de tels systèmes. Il a été proposé un nouvel algorithme de soustraction du fond, sélectif et adaptatif, pour adapter le système à des changements de luminosité et de la structure de la scène. En outre, pour rendre applicable le système à des environnements réels, des heuristiques ont été proposées pour la résolution des différents problèmes : ombres, bruit, etc. Les résultats produits sur la phase de détection d’objets montrent que les techniques proposées sont robustes et utilisables en temps réels grâce à un temps de calcul peu élevé. L’objet principal de la thèse a concerné la phase de suivi d’objets. Dans cette thèse, nous proposons un nouvel algorithme basé sur une expérimentation des objets qui utilisent les pyramides de graphes. Des tests expérimentaux sur des bases de données standard et sur des index attestés pour l’évaluation des algorithmes de suivi d’objets en présence d’occlusions montrent que cette approche est très prometteuse
25
Conte, Donatello Jolion Jean-Michel Vento Mario. "Detection, tracking, and behaviour analysis of moving people in intelligent video surveillance systems a graph based approach /." Villeurbanne : Doc'INSA, 2006. http://docinsa.insa-lyon.fr/these/pont.php?id=conte.
Full text
26
Nam, Do H. "Methodologies for integrating traffic flow theory, ITS and evolving surveillance technologies." Diss., This resource online, 1995. http://scholar.lib.vt.edu/theses/available/etd-06062008-165829/.
Full text
27
Kamiya, Keitaro. "A framework of vision-based detection-tracking surveillance systems for counting vehicles." Thesis, Georgia Institute of Technology, 2012. http://hdl.handle.net/1853/45937.
Full textAbstract:
This thesis presents a framework for motor vehicle detection-tracking surveillance systems. Given an optimized object detection template, the feasibility and effectiveness of the methodology is considered for vehicle counting applications, implementing both a filtering operation of false detection, based on the speed variability in each segment of traffic state, and an occlusion handling technique which considers the unusual affine transformation of tracking subspace, as well as its highly fluctuating averaged acceleration data. The result presents the overall performance considering the trade-off relationship between true detection rate and false detection rate. The filtering operation achieved significant success in removing the majority of non-vehicle elements that do not move like a vehicle. The occlusion handling technique employed also improved the systems performance, contributing counts that would otherwise be lost. For all video samples tested, the proposed framework obtained high correct count (>93% correct counting rate) while simultaneously minimizing the false count rate. For future research, the author recommends the use of more sophisticated filters for specific sets of conditions as well as the implementation of discriminative classifier for detecting different occlusion cases.
28
Schulz, Brian L. P., and Bronchae M. Brown. "The effects of the joint multi-mission electro-optical system on littoral maritime Intelligence, Surveillance, and Reconnaissance operations." Thesis, Monterey, California. Naval Postgraduate School, 2009. http://hdl.handle.net/10945/4638.
Full textAbstract:
Approved for public release, distribution unlimitedThe United States Department of Defense finds itself in a period of reduced resources and growing requirements. In the field of Intelligence, Surveillance, and Reconnaissance (ISR), there have been calls for both manpower and system cuts, while collection requirements continue to increase. One proposed method for maximizing ISR collection efforts is the development of multi-mission capable collection equipment. In support of this concept, BAE Systems has developed the Joint Multi-Mission Electro-optical System (JMMES). Designed for potential use on both manned and unmanned aircraft, JMMES is capable of multi-mission integration and target prosecution without the need to exchange system components or system operator, thus increasing flexibility, responsiveness, and capabilities, while reducing manning and cost requirements. JMMES incorporates multi-spectral technology and advanced search algorithms to enhance autonomous collection capabilities. Our thesis investigates how a JMMES equipped SH-60 variant aircraft affects U.S. ISR capabilities in the littoral regions, specifically in the areas of Anti Submarine Warfare (ASW), Surface Warfare (SUW), Maritime Interdiction Operations (MIO), and Search and Rescue (SAR). We teamed with the faculty research group in conducting JCTD test flights during Trident Warrior 2009. Utilizing both quantitative and qualitative results and analysis from the exercise flights and post-flight surveys, we developed an organizational simulation model, using VDT, to evaluate the benefits of JMMES.
29
Brown, Bronchae M. Schulz Brian L. P. "The effects of the joint multi-mission electro-optical system on littoral maritime Intelligence, Surveillance, and Reconnaissance operations." Monterey, California : Naval Postgraduate School, 2009. http://edocs.nps.edu/npspubs/scholarly/theses/2009/Sep/09Sep%5FBrown.pdf.
Full textAbstract:
Thesis (M.S. in Information Warfare Systems Engineering)--Naval Postgraduate School, September 2009.Thesis Advisor(s): last name, first name ; "September 2009." Description based on title screen as viewed on November 05, 2009. Author(s) subject terms: Infrared, Electro-optic, Joint Capability Technology Demonstration, Intelligence Surveillance and Reconnaissance, Modeling and Simulation Includes bibliographical references (p. 141-144). Also available in print.
30
Younis, Zaki Mohamed. "An ontological approach for monitoring and surveillance systems in unregulated markets." Thesis, University of Manchester, 2013. https://www.research.manchester.ac.uk/portal/en/theses/an-ontological-approach-for-monitoring-and-surveillance-systems-in-unregulated-markets(056f8010-08b2-4eb0-a75d-5301b899ec90).html.
Full textAbstract:
Ontologies are a key factor of Information management as they provide a common representation to any domain. Historically, finance domain has suffered from a lack of efficiency in managing vast amounts of financial data, a lack of communication and knowledge sharing between analysts. Particularly, with the growth of fraud in financial markets, cases are challenging, complex, and involve a huge volume of information. Gathering facts and evidence is often complex. Thus, the impetus for building a financial fraud ontology arises from the continuous improvement and development of financial market surveillance systems with high analytical capabilities to capture frauds which is essential to guarantee and preserve an efficient market.This thesis proposes an ontology-based approach for financial market surveillance systems. The proposed ontology acts as a semantic representation of mining concepts from unstructured resources and other internet sources (corpus). The ontology contains a comprehensive concept system that can act as a semantically rich knowledge base for a market monitoring system. This could help fraud analysts to understand financial fraud practices, assist open investigation by managing relevant facts gathered for case investigations, providing early detection techniques of fraudulent activities, developing prevention practices, and sharing manipulation patterns from prosecuted cases with investigators and relevant users. The usefulness of the ontology will be evaluated through three case studies, which not only help to explain how manipulation in markets works, but will also demonstrate how the ontology can be used as a framework for the extraction process and capturing information related to financial fraud, to improve the performance of surveillance systems in fraud monitoring. Given that most manipulation cases occur in the unregulated markets, this thesis uses a sample of fraud cases from the unregulated markets. On the empirical side, the thesis presents examples of novel applications of text-mining tools and data-processing components, developing off-line surveillance systems that are fully working prototypes which could train the ontology in the most recent manipulation techniques.
31
Le, Mortellec Antoine. "Proposition d'une architecture de surveillance "active" à base d'agents intelligents pour l'aide à la maintenance de systèmes mobiles - Application au domaine ferroviaire." Phd thesis, Université de Valenciennes et du Hainaut-Cambresis, 2014. http://tel.archives-ouvertes.fr/tel-00947981.
Full textAbstract:
Ces deux dernières décennies, les systèmes embarqués ont été introduits dans de nombreux domaines d'application (transport, industrie, habitat, médical...). Ces systèmes se sont vu confier des tâches plus importantes pour délivrer de nouveaux services aux utilisateurs avec des délais de mise sur le marché toujours plus courts et à moindre coût. L'intégration rapide de ces systèmes au sein de produits manufacturés est un avantage concurrentiel pour les industriels. Cependant, les pannes associées à ces systèmes et le niveau de complexité croissant des équipements ont rendu les interventions de maintenance bien plus délicates. L'identification des causes de certaines pannes représente actuellement un véritable challenge dans les activités de la maintenance. Elles entrainent une indisponibilité excessive des équipements.Cette thèse propose une architecture générique de surveillance "active" pour l'aide à la maintenance de systèmes mobiles. Cette architecture repose sur des entités de surveillance "intelligentes" capables d'évaluer l'état de santé des équipements surveillés. Notre contribution se situe à la rencontre de différentes communautés de Recherche et s'appuie notamment sur des concepts développés par lacommunauté PHM (Pronostics and Health Management).L'architecture proposée est mise en œuvre et appliquée a la surveillance d'un système réel de transport ferroviaire dans le cadre du projet SURFER (SURveillance active FERroviaire) conduit par Bombardier-Transport.
32
Brax, Nicolas. "Self-adaptive multi-agent systems for aided decision-making : an application to maritime surveillance." Toulouse 3, 2013. http://thesesups.ups-tlse.fr/2196/.
Full textAbstract:
L'activité maritime s'est fortement développée ces dernières années et sert de support à de nombreuses activités illicites. Il est devenu nécessaire que les organismes impliqués dans la surveillance maritime disposent de systèmes efficaces pour les aider à identifier ces activités illicites. Les Systèmes de Surveillance Maritime doivent observer de manière efficace un espace maritime large, à identifier des anomalies de comportement des navires évoluant dans l'espace en question, et à déclencher des alertes documentées si ces anomalies amènent à penser que les navires ont un comportement suspect. Nous proposons un modèle générique de système multi-agents, que nous appelons MAS4AT, capable de remplir deux des différents rôles d'un système de surveillance : la représentation numérique des comportements des entités surveillées et des mécanismes d'apprentissage pour une meilleure efficacité. MAS4AT est intégré au système I2CThe maritime activity has widely grow in the last few years and is the witness of several illegal activities. It has become necessary that the organizations involved in the maritime surveillance possess efficient systems to help them in their identification. The maritime surveillance systems must observe a wide maritime area, identify the anomalies in the behaviours of the monitored ships et trigger alerts when these anomalies leads to a suspicious behavior. We propose a generic agent model, called MAS4AT, able to fulfil two main roles of a surveillance system: the numerical representation of the behaviours of the monitored entities and learning mechanisms for a better efficiency. MAS4AT is integrated in the system I2C
33
Davis, Cledo L. "The systems integration of autonomous behavior analysis to create a "Maritime Smart Environment" for the enhancement of maritime domain awareness." Thesis, Monterey, California : Naval Postgraduate School, 2010. http://edocs.nps.edu/npspubs/scholarly/theses/2010/Jun/10Jun%5FDavis.pdf.
Full textAbstract:
Thesis (M.S. in Systems Engineering)--Naval Postgraduate School, June 2010.Thesis Advisor(s): Goshorn, Rachel ; Goshorn, Deborah. "June 2010." Description based on title screen as viewed on June 24, 2010. Author(s) subject terms: Anomaly Detection, Artificial Intelligence, Automation, Behavior Analysis, Distributed Artificial Intelligence, Intelligence-Surveillance-Reconnaissance, Maritime Domain Awareness, Maritime Force Protection, Multi-agent Systems, Network-centric Operations, Network-centric Systems Engineering, Network-centric Warfare, Smart Sensor Networks, Systems Engineering, Systems Integration, System of Systems. Includes bibliographical references (p. 209-212). Also available in print.
34
Yu, Shen. "A Bayesian machine learning system for recognizing group behaviour." Thesis, McGill University, 2009. http://digitool.Library.McGill.CA:8881/R/?func=dbin-jump-full&object_id=32565.
Full text
35
Голінка, А. Ю. "Інтелектуальна система керування автомобільною стоянкою." Master's thesis, Сумський державний університет, 2020. https://essuir.sumdu.edu.ua/handle/123456789/79551.
Full textAbstract:
Проведено аналіз типів автоматичних систем автостоянок, порівняльний аналіз існуючих методів
виявлення контуру об’єкту, розроблений функціонал автоматичного виявлення заповнення
автомобільної парковки, проведено тестування реалізації, зроблені висновки по результатам
тестування, дана оцінка актуальності та ефективності даної системи.
36
Guastella, Davide Andrea. "Dynamic learning of the environment for eco-citizen behavior." Thesis, Toulouse 3, 2020. http://www.theses.fr/2020TOU30160.
Full textAbstract:
Le développement de villes intelligentes et durables nécessite le déploiement des technologies de l'information et de la communication (ITC) pour garantir de meilleurs services et informations disponibles à tout moment et partout. Comme les dispositifs IoT devenant plus puissants et moins coûteux, la mise en place d'un réseau de capteurs dans un contexte urbain peut être coûteuse. Cette thèse propose une technique pour estimer les informations environnementales manquantes dans des environnements à large échelle. Notre technique permet de fournir des informations alors que les dispositifs ne sont pas disponibles dans une zone de l'environnement non couverte par des capteurs. La contribution de notre proposition est résumée dans les points suivants : - limiter le nombre de dispositifs de détection à déployer dans un environnement urbain ; - l'exploitation de données hétérogènes acquises par des dispositifs intermittents ; - le traitement en temps réel des informations ; - l'auto-calibration du système. Notre proposition utilise l'approche AMAS (Adaptive Multi-Agent System) pour résoudre le problème de l'indisponibilité des informations. Dans cette approche, une exception est considérée comme une situation non coopérative (NCS) qui doit être résolue localement et de manière coopérative. HybridIoT exploite à la fois des informations homogènes (informations du même type) et hétérogènes (informations de différents types ou unités) acquises à partir d'un capteur disponible pour fournir des estimations précises au point de l'environnement où un capteur n'est pas disponible. La technique proposée permet d'estimer des informations environnementales précises dans des conditions de variabilité résultant du contexte d'application urbaine dans lequel le projet est situé, et qui n'ont pas été explorées par les solutions de l'état de l'art : - ouverture : les capteurs peuvent entrer ou sortir du système à tout moment sans qu'aucune configuration particulière soit nécessaire ; - large échelle : le système peut être déployé dans un contexte urbain à large échelle et assurer un fonctionnement correct avec un nombre significatif de dispositifs ; - hétérogénéité : le système traite différents types d'informations sans aucune configuration a priori. Notre proposition ne nécessite aucun paramètre d'entrée ni aucune reconfiguration. Le système peut fonctionner dans des environnements ouverts et dynamiques tels que les villes, où un grand nombre de capteurs peuvent apparaître ou disparaître à tout moment et sans aucun préavis. Nous avons fait différentes expérimentations pour comparer les résultats obtenus à plusieurs techniques standard afin d'évaluer la validité de notre proposition. Nous avons également développé un ensemble de techniques standard pour produire des résultats de base qui seront comparés à ceux obtenus par notre proposition multi-agentsThe development of sustainable smart cities requires the deployment of Information and Communication Technology (ICT) to ensure better services and available information at any time and everywhere. As IoT devices become more powerful and low-cost, the implementation of an extensive sensor network for an urban context can be expensive. This thesis proposes a technique for estimating missing environmental information in large scale environments. Our technique enables providing information whereas devices are not available for an area of the environment not covered by sensing devices. The contribution of our proposal is summarized in the following points: * limiting the number of sensing devices to be deployed in an urban environment; * the exploitation of heterogeneous data acquired from intermittent devices; * real-time processing of information; * self-calibration of the system. Our proposal uses the Adaptive Multi-Agent System (AMAS) approach to solve the problem of information unavailability. In this approach, an exception is considered as a Non-Cooperative Situation (NCS) that has to be solved locally and cooperatively. HybridIoT exploits both homogeneous (information of the same type) and heterogeneous information (information of different types or units) acquired from some available sensing device to provide accurate estimates in the point of the environment where a sensing device is not available. The proposed technique enables estimating accurate environmental information under conditions of uncertainty arising from the urban application context in which the project is situated, and which have not been explored by the state-of-the-art solutions: * openness: sensors can enter or leave the system at any time without the need for any reconfiguration; * large scale: the system can be deployed in a large, urban context and ensure correct operation with a significative number of devices; * heterogeneity: the system handles different types of information without any a priori configuration. Our proposal does not require any input parameters or reconfiguration. The system can operate in open, dynamic environments such as cities, where a large number of sensing devices can appear or disappear at any time and without any prior notification. We carried out different experiments to compare the obtained results to various standard techniques to assess the validity of our proposal. We also developed a pipeline of standard techniques to produce baseline results that will be compared to those obtained by our multi-agent proposal
37
Arantes, Júnior Wilmondes Manzi de. "P. A. S. Pluggable Alert System : un système pour la génération et l'affichage d'alertes médicales adaptées à l'utilisateur." Lyon, INSA, 2006. http://theses.insa-lyon.fr/publication/2006ISAL0025/these.pdf.
Full textAbstract:
Ce travail concerne la conception et le développement d’un système de détection d’alertes médicales adaptables au contexte de consultation dans le cadre des réseaux de soins. Ces alertes sont créées à l’aide de variables linguistiques associées à des niveaux d’importance (alerte si âge = âgé; important et température = très-chaud; très-important) et dont les rapports d’influence (le poids dépend de l’âge) sont modélisés par un graphe orienté pondéré. Chaque alerte déclenchée présente deux indices de qualité – dits d’applicabilité et de confiance – qui indiquent dans quelle mesure le patient est concerné et jusqu’à quel point elle est digne de confiance. Notre système est aussi capable de traiter de façon transparente les informations manquantes en s’appuyant sur une base historique utilisée pour prédire le valeurs inconnues. Au niveau de l’affichage, un module multi-agents se charge d’adapter les alertes déclenchées au contexte, qui est représenté, entre autres, par les caractéristiques du professionnel de santé, le dispositif d’affichage et l’urgence. L’adaptation est alors menée par trois agents intelligents – le patient, le médecin et l’alerte – qui négocient sur la qualité requise pour chaque dimension de l’interface finale : le contenu, le graphisme et la sécurité. Ensuite, des appels d’offre correspondants sont diffusés dans trois sociétés d’agents de service exécuteurs de tâches qui représentant ces trois dimensions et ceux qui les gagnants collaborent pour construire l’interface de l’alerte. Finalement, les tests réalisés sur le module de détection – qui fera l’objet d’un dépôt de brevet – se sont montrés très satisfaisantsWe propose a system that is able to detect and trigger user-defined medical alerts in the context of healthcare networks. Such alerts are created by using fuzzy linguistic variables associated with importance levels (e. G. Alert if age = elderly; important and air-temperature = very-hot; very-important) and whose dependency relationships (e. G. The weight depends on the age) are modeled through a weighted oriented graph. Each alert the system triggers has two quality indicators – an applicability level and a trust level – which state to which extent the patient is concerned and how reliable it is. Our system is also able to transparently infer missing information by using an historical database containing previously reported similar cases. Finally, a multi-agents display module adapts each alert to the context, which is represented by the patient (elderly, etc. ), the healthcare user (physician, etc. ), the display device (PC, PDA, etc. ), the place (hospital, etc. ) and the urgency of the alert itself (very urgent, etc. ). The adaptation process is controlled by three intelligent agents – the patient, the physician and the alert – which negotiate to agree on the min-max quality levels required for the three dimensions of display: contents (information that will be shown), graphics (graphic components that will be used) and security (protection mechanisms to use). Then, the corresponding task announcements are broadcasted within three societies of reactive agents (which have not cognitive capabilities and simply perform tasks) representing these dimensions and the winning ones collaborate to build the interface of the alert. The final system will be integrated into the hospital information system provided by the company that has sponsored my research and will be patented as soon as possible
38
Crouch, Collier Craig. "Integration of mini-UAVs at the tactical operations level : implications of operations, implementation, and information sharing /." Thesis, Monterey, Calif. : Springfield, Va. : Naval Postgraduate School ; Available from National Technical Information Service, 2005. http://library.nps.navy.mil/uhtbin/hyperion/05Jun%5FCrouch.pdf.
Full text
39
Jeveme, Panta Franck. "Modélisation des métadonnées multi sources et hétérogènes pour le filtrage négatif et l'interrogation intelligente de grands volumes de données : application à la vidéosurveillance." Thesis, Toulouse 3, 2020. http://www.theses.fr/2020TOU30098.
Full textAbstract:
En raison du déploiement massif et progressif des systèmes de vidéosurveillance dans les grandes métropoles, l'analyse a posteriori des vidéos issues de ces systèmes est confrontée à de nombreux problèmes parmi lesquels: (i) l'interopérabilité, due aux différents formats de données (vidéos) et aux spécifications des caméras propres à chaque système ; (ii) le grand temps d'analyse lié à l'énorme quantité de données et métadonnées générées ; et (iii) la difficulté à interpréter les vidéos qui sont parfois à caractère incomplet. Face à ces problèmes, la nécessité de proposer un format commun d'échange des données et métadonnées de vidéosurveillance, de rendre le filtrage et l'interrogation des contenus vidéo plus efficaces, et de faciliter l'interprétation des contenus grâce aux informations exogènes (contextuelles) est une préoccupation incontournable. De ce fait, cette thèse se focalise sur la modélisation des métadonnées multi sources et hétérogènes afin de proposer un filtrage négatif et une interrogation intelligente des données, applicables aux systèmes de vidéosurveillance en particulier et adaptables aux systèmes traitant de grands volumes de données en général. L'objectif dans le cadre applicatif de cette thèse est de fournir aux opérateurs humains de vidéosurveillance des outils pour les aider à réduire le grand volume de vidéo à traiter ou à visionner et implicitement le temps de recherche. Nous proposons donc dans un premier temps une méthode de filtrage dit "négatif", qui permet d'éliminer parmi la masse de vidéos disponibles celles dont on sait au préalable en se basant sur un ensemble de critères, que le traitement n'aboutira à aucun résultat. Les critères utilisés pour l'approche de filtrage négatif proposé sont basés sur une modélisation des métadonnées décrivant la qualité et l'utilisabilité/utilité des vidéos. Ensuite, nous proposons un processus d'enrichissement contextuel basé sur les métadonnées issues du contexte, et permettant une interrogation intelligente des vidéos. Le processus d'enrichissement contextuel proposé est soutenu par un modèle de métadonnées extensible qui intègre des informations contextuelles de sources variées, et un mécanisme de requêtage multiniveaux avec une capacité de raisonnement spatio-temporel robuste aux requêtes floues. Enfin, nous proposons une modélisation générique des métadonnées de vidéosurveillance intégrant les métadonnées décrivant le mouvement et le champ de vue des caméras, les métadonnées issues des algorithmes d'analyse des contenus, et les métadonnées issues des informations contextuelles, afin de compléter le dictionnaire des métadonnées de la norme ISO 22311/IEC 79 qui vise à fournir un format commun d'export des données extraites des systèmes de vidéosurveillance. Les expérimentations menées à partir du framework développé dans cette thèse ont permis de démontrer la faisabilité de notre approche dans un cas réel et de valider nos propositionsDue to the massive and progressive deployment of video surveillance systems in major cities, a posteriori analysis of videos coming from these systems is facing many problems, including the following: (i) interoperability, due to the different data (video) formats and camera specifications associated to each system; (ii) time-consuming nature of analysis due to the huge amount of data and metadata generated; and (iii) difficulty to interpret videos which are sometimes incomplete. To address these issues, the need to propose a common format to exchange video surveillance data and metadata, to make video content filtering and querying more efficient, and to facilitate the interpretation of content using external (contextual) information is an unavoidable concern. Therefore, this thesis focuses on heterogeneous and multi-source metadata modeling in order to propose negative filtering and intelligent data querying, which are applicable to video surveillance systems in particular and adaptable to systems dealing with large volumes of data in general. In the applicative context of this thesis, the goal is to provide human CCTV operators with tools that help them to reduce the large volume of video to be processed or viewed and implicitly reduce search time. We therefore initially propose a so-called "negative" filtering method, which enables the elimination from the mass of available videos those that it is know in advance, based on a set of criteria, that the processing will not lead to any result. The criteria used for the proposed negative filtering approach are based on metadata modeling describing video quality and usability/usefulness. Then, we propose a contextual enrichment process based on metadata from the context, enabling intelligent querying of the videos. The proposed contextual enrichment process is supported by a scalable metadata model that integrates contextual information from a variety of sources, and a multi-level query mechanism with a spatio-temporal reasoning ability that is robust to fuzzy queries. Finally, we propose a generic metadata modeling of video surveillance metadata integrating metadata describing the movement and field of view of cameras, metadata from content analysis algorithms, and metadata from contextual information, in order to complete the metadata dictionary of the ISO 22311/IEC 79 standard, which aims to provide a common format to export data extracted from video surveillance systems. The experiments performed using the framework developed in this thesis showed the reliability of our approach in a real case and enabled the validation of our proposals
40
Khan, Kamran. "Refractive conditions in Arabian Sea and their effects on ESM and airborne radar operations." Thesis, Monterey, California : Naval Postgraduate School, 1990. http://handle.dtic.mil/100.2/ADA238273.
Full textAbstract:
Thesis (M.S. in Systems Engineering (Electronic Warfare))--Naval Postgraduate School, September 1990.Thesis Advisor(s): Davidson, Kenneth L. ; Powell, James R. "September 1990." Description based on title screen as viewed on December 29, 2009. DTIC Descriptor(s): Frequency, Electronic Warfare, Aircraft, Airborne, Electronic Equipment, Microwave Equipment, Radar, Profiles, Ducts, Meteorology, Communication And Radio Systems, Refraction, Arabian Sea, Refractometers, Military Operations. DTIC Identifier(s): Radar interference, meteorological phenomena, theses. Author(s) subject terms: Refractivity, Arabian Sea refractive conditions, ESM airborne radar, airborne microwave refractometer (AMR), IREPS, EREPS. Includes bibliographical references (p. 96-97). Also available in print.
41
GALDELLI, ALESSANDRO. "Applied Artificial Intelligence for Precision Fishing: identification and classification of fishing activities." Doctoral thesis, Università Politecnica delle Marche, 2021. http://hdl.handle.net/11566/289710.
Full textAbstract:
Il costante incremento delle attività di pesca e del traffico marittimo in generale hanno reso il monitoraggio e la classificazione delle attività delle navi una sfida aperta nello scenario marino. Il continuo sfruttamento delle risorse ittiche ha ridotto drasticamente l’abbondanza di queste risorse con conseguenze negative sullo stesso settore della pesca. Nel corso degli anni sono stati introdotti degli strumenti che inizialmente però erano impiegati solamente per migliorare la sicurezza del traffico marittimo. La necessità di risolvere il problema del monitoraggio e della classificazione delle attività delle navi nella nuova era dell’Intelligenza Artificiale (IA) ha portato allo sviluppo e all’applicazione di nuovi metodi nel campo del Machine Learning (ML). In particolare, l’applicazione della IA in questo contesto definisce un nuovo concetto che prende il nome di Precision Fishing. Il lavoro svolto in questa tesi è stato sviluppato in collaborazione con l’Istituto per le Risorse Biologiche e le Biotecnologie Marine del CNR (CNR-IRBIM). L’obiettivo della tesi è incrementare il controllo delle attività di pesca analizzando dati provenienti dal Sistema di Identificazione Automatica (AIS) ed integrandoli ad esempio con le immagini satellitari “Synthetic Aperture RADAR” (SAR). Gli obiettivi della presente tesi hanno riguardato (i) l’identificazione e (ii) la classificazione delle attività di pesca e (iii) l’individuazione delle attività illegali, non dichiarate e non regolamentate (INN), mediante approcci di Intelligenza Artificiale. Nel primo tema di ricerca si presenta un algoritmo in grado di individuare ogni singola sessione di pesca, ossia tutto ciò che accade da quando la nave lascia il porto di partenza fino al porto di destinazione. Per ottenere questo risultato, la prima operazione svolta è quella del filtraggio degli outliers (dati AIS anomali o errati), che è stato ottenuto grazie ad un processo di interpolazione. L’algoritmo sviluppato utilizza un insieme di regole per identificare ciascuna sessione di pesca. Un altro aspetto innovativo dell’algoritmo rispetto allo stato dell’arte è quello di ricostruire sessioni di pesca incomplete, ossia quando quest’ultime non hanno una distribuzione temporalmente uniforme dei dati AIS. L’affidabilità del metodo proposto è stata valutata su un dataset validato da esperti nel settore, ed i risultati ottenuti hanno dimostrato che l’efficacia del metodo è superiore rispetto allo stato dell’arte. Nella seconda tematica si propone una serie di algoritmi basati sulle tecnologie dell’IA al fine di classificare le attività di pesca. In dettaglio vengono realizzati diversi algoritmi di classificazione utilizzando diverse tecniche di Machine Learning e di Deep Learning. L’innovazione apportata allo stato dell’arte rispetto agli obiettivi sopra riportati è nello sviluppo di algoritmi basati su IA che automatizzano processi di analisi dati per supportare decision maker nell’ambito della Precision Fishing. L’affidabilità dei metodi proposti è stata indagata utilizzando dataset validati da esperti nel settore e dallo studio dei comportamenti delle navi su diversi anni. I risultati ottenuti sono superiori allo stato dell’arte e questo fa si che alcuni algoritmi proposti si candidano ad essere considerati come gold standard. Nel terzo tema di ricerca si presenta un algoritmo per l’identificazione delle attività di pesca INN. In questo caso l’utilizzo del solo sistema AIS è insufficiente perché nella maggior parte dei casi, quando la nave è impegnata in questo tipo di attività, i sistemi di bordo vengono spenti in modo da non poter essere rintracciati. La soluzione proposta è quella di integrare i dati AIS con le immagini satellitari SAR in modo da ricostruire l’informazione mancante, e grazie all’algoritmo di classificazione delle attività di pesca, vengono rilevate tutte quelle che sono ritenute sospette. Il metodo proposto è stato validato da esperti nel settore e dall’analisi dei registri di bordo integrando la conoscenza dei sistemi di pesca.The constant increasing of fishing activities and marine traffic have made the monitoring and the classification of the ships activities an open challenge in marine scenario. Continued exploitation of fish resources has drastically reduced the abundance of these resources, with negative consequences on the fisheries sector itself. Over the years, some tools have been introduced, but initially they were only used to improve the safety of maritime traffic. The necessity of solving the problem of the monitoring and the classification of the ships activities in the new era of Artificial Intelligence (AI) leads to the development and to the implementation of new methods in Machine Learning (ML). In particular, the application of AI in this context defines a new concept called Precision Fishing. The work of this thesis has been developed in collaboration with “Istituto per le Risorse Biologiche e le Biotecnologie Marine” of the CNR (CNR-IRBIM). The aim of this research is to increase fisheries control by analysing Automatic Identification System (AIS) data and integrating them with additional data such as “Synthetic Aperture RADAR” (SAR) images. The objectives of this thesis regarded (i) the identification and (ii) the classification of fishing activities; (iii) the identification of illegal, unreported and unregulated (IUU) fishing activities through AI approaches. In the first topic, it is described an algorithm able to identify every single fishing session, meaning everything that happens from when the ship leaves the port of departure to the port of destination. In order to obtain this result, the first operation carried out is the filtering of outliers (on-land or erroneous AIS data), which has been achieved through a process of interpolation. The algorithm developed uses a rule set to identify each fishing session. Another innovative aspect of the algorithm compared to the state of the art is that it reconstructs incomplete fishing sessions, meaning those that do not have a temporally uniform distribution of AIS data. The reliability of the proposed method was evaluated on a dataset validated by experts in the field, and the results obtained showed that the effectiveness of the method outperformed the state of the art. In the second research topic, it is proposed a set of algorithms based on AI technologies in order to classify fishing activities. In detail, several classification algorithms are implemented using different Machine Learning and Deep Learning techniques. The innovation of this thesis over the state of the art is the design and the development of AI algorithms to support decision makers in the Precision Fishing field using AIS and satellite data. The reliability of the proposed methods was investigated using datasets validated by experts in the field and by studying the behaviour of ships over the years. The results obtained are better than the state of the art and this makes some of the proposed algorithms candidates to be considered as gold standard. In the third topic, it is presented an algorithm for the identification of IUU fishing activities. In this case the use of the AIS system alone is insufficient because in most cases, when the ship is engaged in this type of activity, the on-board systems are switched off so that the vessel cannot be located. The solution proposed is to integrate AIS data with SAR satellite images in order to recover the missing information, and thanks to the classification of fishing activities algorithm all those that are considered suspicious are detected. The proposed method has been validated by experts in the field and by the analysis of logbooks integrating knowledge of fishing systems.
42
Wu, Dong-Yan, and 吳東諺. "Intelligent House Surveillance System." Thesis, 2009. http://ndltd.ncl.edu.tw/handle/69882990437609533666.
Full textAbstract:
碩士國立勤益科技大學
電子工程系
97
Because of the trend toward more concentrated population in the cities, people pay more attention to the environment of house and the quality of life than before. For these reasons, this study focus on some issues, including the leak of gas, the indoor temperature, the humidity, the indoor brightness and the home-intruder. The study mainly use SPARTAN3 TEST BOARD and internet, adding on temperature sensor, humidity sensor, vibration sensor, gas sensor, photosensitive resistance and the human body sensor. The results will be real-time displayed on the testing board and on the homepage, by those we can make real-time monitoring of environment of house true.
43
Chang, Hung-Hsiang, and 張宏祥. "Intelligent Green Energy Surveillance System." Thesis, 2011. http://ndltd.ncl.edu.tw/handle/65039323944956623899.
Full textAbstract:
碩士淡江大學
電機工程學系碩士在職專班
99
Wireless broadband transformation is making broadband services anywhere, any time and in any form a reality today. Recently a university in Taiwan was experiencing dramatic power usage increases due to its growing number of campus buildings and students. Aiming to analyze their power consumption and increase their power efficiency across buildings, the university wanted to build a power management system utilizing wireless hardware and software. The main purpose of this thesis is to propose a infrastructure (or scheme) of energy conservation management, control, and monitor system based on wireless communication technology. To achieve energy conservation efficiently, a micro-controller can control the wireless switches based on the condition of surrounding area to avoid wasting of energy by human ignorance. Currently, this research is using AM frequency 433.92 as wireless media with frequency relay to extend the range of communication. The condition of monitoring area will be transmitting to a PC in a control center for further processing. The monitoring PC can display condition of area, adjust setting, monitor state of switches, and analyse received information. Also, the PC can control switches based on analysed information and predetermined setting to have most efficient power consumption. We asks for people to save a safe, low polluted and low energy consumption living environment for the world and the coming generation. Hoping everyone can gradually expand influence to family, friends, society and the whole world.
44
Chen, Shih-wei, and 陳師偉. "Architecture Design for Intelligent Surveillance System." Thesis, 2016. http://ndltd.ncl.edu.tw/handle/55259860404743267403.
Full textAbstract:
碩士國立中央大學
電機工程學系
104
The digital surveillance system becomes more and more popular in recent years. It attempts to raise amount of high resolution cameras, consequently those systems stupendously increase the computational load on central server. As in the intelligent object recognition processing flow, the technique on segmentation and tracking multiple targets, such as tracking group of people through occlusion is still challenging. In this paper, we present an architecture design for intelligent surveillance system. Mainly made up of four image processing module composed, contains foreground detection, sliced connected component labeling, object grouping and object tracking. We have a complete system-level solution on algorithm and VLSI implementation. This design is using TSMC 90 nm library with 4 MHz operation frequency. Without calculating memory of gate count about 18.71K. Power consumption about 11.4037mW and memory usage is 92.288Kbytes. Simply use the center and boundary box of the object will be able to track objects, and solve the problem occurs when occlusion.
45
Ku, Min-Yu, and 古閔宇. "Intelligent Video Surveillance and Recognition System." Thesis, 2009. http://ndltd.ncl.edu.tw/handle/06468539799880242017.
Full textAbstract:
博士國防大學理工學院
國防科學研究所
97
Capabilities of a video-based monitoring and recognizing system can be applied to many categories, for example vision-based vehicle detection system of traffic management, intrusion detection system of security and face recognition access control system of biometrics etc. This thesis proposes an algorithm with a low-cost camera by Taiwan to replace high-priced imports of vision-based vehicle detection system. We propose an algorithm to extract initial color backgrounds from surveillance videos using a probability-based background extraction algorithm. With the proposed algorithm, the initial background can be extracted accurately and quickly, while using relatively little memory. The intrusive objects can then be segmented quickly and correctly by a robust object segmentation algorithm. The segmentation algorithm analyzes the threshold values of the background subtraction from the prior frame to obtain good quality and update while minimizing execution time and maximizing detection accuracy. The segmentation and recognition method uses the length, width, and roof size to classify vehicles, even when occlusive vehicles are continuously merging from one frame to the next. The segmented objects can be recognized and counted in accordance with their varying features, via the proposed recognition and tracking methods. The color background images can be extracted efficiently and quickly from color image sequences and updated in real time to overcome any variation in illumination conditions. Experimental results for various environmental sequences and weather conditions are provided to demonstrate the robustness, accuracy, effectiveness, and memory economy of the proposed algorithm.
46
Chen, Shen-Chi, and 陳宣輯. "Vision Sensing Techniques for Intelligent Surveillance System." Thesis, 2016. http://ndltd.ncl.edu.tw/handle/13968549284508916009.
Full textAbstract:
博士國立臺灣大學
資訊工程學研究所
104
With the development of intelligent surveillance systems, video analysis, and recognition technology have become the most important core techniques in this field. In order to construct a surveillance system with higher intelligence, this research proposes a number of advanced video recognition technologies, including the camera interference/tampering detection, pedestrian detection, abandoned luggage detection, pedestrian re-identification and intelligent interface for visualization. Video surveillance uses cameras as the primary input sensor to achieve automatic monitoring. Therefore, how to protect the camera has become the top priority. We propose real-time camera sabotage/tampering detection technology which quickly detects whether or not cameras are hindered by deliberate shelter, disorientation, out of focus, disconnection and other damage via the video analysis. We initially locate the key points whose appearances are relatively stable. Monitoring the changes of these key points and scene structure can detect the tampering events precisely and efficiently. Our method requires lower computational cost and obtains higher stability and accuracy rate in comparison to the existing methods. After protecting cameras, we propose a scene-specific pedestrian detection and object classification. Our approach is location-based, which cab discover scene-dependent discriminative features to identifying foreground objects of different categories (e.g., pedestrians, bicycles, and vehicles). We incorporate a similarity grouping procedure capable of gathering more consistent training examples from a considerably larger neighbor area and train the specific pedestrian detectors for each grouped local area. Our approach gets significant improvement in detection and classification comparing the traditional generic object detector and classifier. Also, we propose an ensemble of invariant features (EIF), which can properly handle the color variations and human poses/viewpoints for matching pedestrian images observed in different cameras. Our proposed method belongs the direct method, which requires no domain learning. The novel features combined both the holistic and region-based features. The holistic features are extracted by using a publicly available pre-trained deep convolutional neural network (DCNN) used in generic object classification. In contrast, the region-based features are extracted based on our proposed two-way Gaussian Mixture Model fitting (2WGMMF), which overcomes the self-occlusion and poses variations. In addition to the appearance feature, the face information is undoubtedly the indispensable vital in video surveillance. We propose a 3D face alignment algorithm in the 2D image based on Active Shape Model. We off-line train a 3D shape model with different view-based local texture models from a 3D database, and then on-line fit a face in a 2D image by these models. This method mainly leverages additional depth information on the traditional 2D image alignment problem and gets a promising improvement compared to the existing model-based and regression-based approaches. Since the human poses, and their gaze directions are especially valuable information to the surveillance system, the head poses can be directly estimated by the alignment result of the proposed 3D model subsequently. Based on the robust pedestrian detection and re-identification algorithm, we also focus the problem of event detection in surveillance cameras. We take the abandoned luggage detection as an example since it is one of the most critical and challenge problems in video surveillance. We propose the complementary background model which combines short- and long-term background models to classify each pixel as 2-bit code where each bit represents a foreground or background. Subsequently, we introduce a finite-state machine framework to identify static foreground regions based on the temporal transition of code patterns and to determine whether the selected area contain abandoned objects by analyzing the back-traced trajectories of luggage owners. The experimental results obtained based on video images from 2006 Performance Evaluation of Tracking and Surveillance (PETS2006), 2007 Advanced Video, Signal-based Surveillance (AVSS2007) databases and NTU data set collected by ourselves. We show that the proposed approach is useful for detecting abandoned luggage and that it outperforms previous methods. Finally, based on the above core technologies, we also propose two advanced visualization interface, which facilitates people to observe quickly and search incidents of pedestrians within a camera network.
47
Jer, Huang Kuen, and 黃堃哲. "An Intelligent Surveillance System with Object Motion Detection." Thesis, 2005. http://ndltd.ncl.edu.tw/handle/20067948954936983036.
Full textAbstract:
碩士長庚大學
電機工程研究所
93
The purpose of the paper is to design an embedded surveillance system for object motion detection. The research was divided into two major parts. One is the embedded system. The other part is image process. A suitable development platform should be selected before the construction of an embedded system. After the selection procedure, an Embedded Linux was compiled based on users’ demand in order to manage peripherals. The process of the compiling assisted in synthesizing the architecture of the operating system. The drivers were chosen according the system hardware. Next, user’s application program was written according to the function of the system. The final process of the first part was to make the operating system and the application program into image files and burn them to our development kit. In image pre-process, we used Video4Linux to get the image information from the camera and calculated adaptive threshold by means of Gaussian Mixture Model. The relation of the background difference image, time difference image and comparative difference image helped find object motion mask image and draw moving objects in rectangle by region segmentation. In terms of background image, appropriate judgment to update region background facilitated more accurate detection of the system on moving object. Finally, two parts of the research were integrated to complete an independent surveillance system for object motion detection.
48
Chang-Ting, Tsai, and 蔡彰庭. "Entrance Guard Surveillance Gearing System Using Intelligent Network." Thesis, 2010. http://ndltd.ncl.edu.tw/handle/22613604365997523990.
Full textAbstract:
碩士國立高雄應用科技大學
電子工程系
98
The present paper proposed the Entrance Guard Surveillance Gearing System Using Intelligent Network, It provides the video recording the entrance and image return of various environment real-time recording. Furthermore, the continuing of video recording the entrance guard system includes intrusion, following and intercom. We can execute these functions by the browser to monitor the home, building, office and work area.The main purpose of this research are discriminating the entrance guard events and controlling and handling each real-time entrance guard condition. In order to achieve the entrance guard management, the system provides the real-time video recording for future analysis. The study can be proved by practicing through the certain large industry in Mailiao. The system can monitor and manage effectively and to reduce the human resource.
49
Huang, Jing-Shian, and 黃景賢. "Cloud-based Intelligent Surveillance System for Faces Detection." Thesis, 2013. http://ndltd.ncl.edu.tw/handle/vpxu86.
Full textAbstract:
碩士崑山科技大學
電機工程研究所
101
Traditional surveillance systems are mostly people need to sit staring at the screen in front of monitoring terminal. Once a long time for staring at the screen, attention will be relatively non-centralized, and it will cause the situation happened that even objects into the monitor screen will be difficult to find, and then cause the emergence of a lot of contingencies or social cases. Therefore this thesis proposed a cloud-based intelligence surveillance system for face detection. In this system, the face detecting function of remote monitoring system can be achieved through the high computing ability of the cloud to reduce the shortcomings of manual monitoring. In face detection, this research uses the Apache software developed by the company Hadoop, the reason for using this software is that Hadoop belong to a distributed computing platform,which can effectively process a large quantity of data through the Map/Reduce parallel processing mode, and the computing result is more fast than the traditional operation way, so it is very helpful for the face image processing speed. The purpose of this thesis is to developa cloud-centric intelligent monitoring system. This cloud system can be divided into three parts. The first part is the CCD(Charge-Coupled-Device) camera control side, which is responsible for transmitting the image data. The second part is the image processing end, which is responsible for the image data computing of face detection, and the treated face data will be remained on the server. Finally, the third part is the network application server, which is used to display the face detecting results.
50
Chang, Cheng-Yu, and 張証喻. "Development of Real-time Intelligent Unmanned Surveillance System." Thesis, 2008. http://ndltd.ncl.edu.tw/handle/6daneq.
Full textAbstract:
碩士國立臺北科技大學
自動化科技研究所
96
Generally speaking, digital image processing technology includes Image Processing and Image Analysis. Object Extraction and Background Separation are two main topics of Image Processing. Moving Edge Detection is generally used to improve the problems of Static Background Subtraction & Temporal Differencing. To solve the problems of Moving Edge Detection, Double Image Separation Technique is proposed in this study. By applying, Object Tracking, Trajectory Recording, Filter, Morphological, and Adjacency, the perfect image can be obtained. In the thesis, a smart Surveillance System by combining digital image processing and Video Understanding has been developed. The system can be used to capture the people in some particular places and to analyze their motions. In case of emergencies, the system will connect to the police automatically and save all data for future application.