Thèses sur le sujet « Integrated surveillance system »

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

Thesis (M.S. in Operations Research)--Naval Postgraduate School, June 2007.
Thesis Advisor(s): Whitaker, Lyn. "June 2007." Description based on title screen as viewed on August 15, 2007. Includes bibliographical references (p. 193-197). Also available in print.

The problem of Third Party Interference (TPI) on Oil and Gas Pipelines is on the rise across the world. TPI is not only common in developing countries but is now occasionally experienced in developed countries including Germany and the UK. The risks posed by these third-party activities on Oil and Gas pipelines are enormous and could be measured in terms of financial costs, environmental damages as well as health and safety implications. The quest for an end to these malicious activities has triggered a lot of studies into the root causes of pipeline TPI, other causes of pipeline failure, risks associated with pipeline failure and their mitigation measures. However, despite the significance of the effects of TPI, very little has been done to proffer an enduring solution through research. This research therefore aims at developing a framework for integrated oil and gas pipeline monitoring and incident mitigation system through integration of various wireless sensors for effective monitoring of oil and gas pipelines. Having identified the existing gaps in literature as lack of reliable, accurate and standard method for oil and gas pipeline risk assessment model, the study undertook a quantitative approach to develop an effective Quantitative Risk Assessment (QRA) model for pipelines. The QRA model developed benchmarks pipeline risk assessment and gives the parameters with which standard QRA could be measured. The research findings indicate that risk associated with Nigerian Pipeline system is in the intolerable region whereas TPI is an increasing menace across the globe. Further findings show that Support Vector Machine (SVM) gave the best performance with 91.2% accuracy while Neural Networks (NN) and Decision Tree (DT) gave 63% and 57% accuracies respectively in terms of pipeline failure mode prediction accuracies. It was recommended that operators should draw out Pipeline Integrity Management (PIM) programs and store pipeline data in a format that captures number of fatalities, property damages and costs as well as volume of oil or gas spilled to ensure that accurate data is obtainable for improved PIM. In conclusion, having achieved its aim and objectives evidenced by the framework, model developed, and the recommendations presented, the research has contributed in no small measure to providing a solution to pipeline incidences.

The incidence of many arboviral diseases is largely associated with social and environmental conditions. Ross River virus (RRV) is the most prevalent arboviral disease in Australia. It has long been recognised that the transmission pattern of RRV is sensitive to socio-ecological factors including climate variation, population movement, mosquito-density and vegetation types. This study aimed to assess the relationships between socio-environmental variability and the transmission of RRV using spatio-temporal analytic methods. Computerised data files of daily RRV disease cases and daily climatic variables in Brisbane, Queensland during 1985-2001 were obtained from the Queensland Department of Health and the Australian Bureau of Meteorology, respectively. Available information on other socio-ecological factors was also collected from relevant government agencies as follows: 1) socio-demographic data from the Australia Bureau of Statistics; 2) information on vegetation (littoral wetlands, ephemeral wetlands, open freshwater, riparian vegetation, melaleuca open forests, wet eucalypt, open forests and other bushland) from Brisbane City Council; 3) tidal activities from the Queensland Department of Transport; and 4) mosquito-density from Brisbane City Council. Principal components analysis (PCA) was used as an exploratory technique for discovering spatial and temporal pattern of RRV distribution. The PCA results show that the first principal component accounted for approximately 57% of the information, which contained the four seasonal rates and loaded highest and positively for autumn. K-means cluster analysis indicates that the seasonality of RRV is characterised by three groups with high, medium and low incidence of disease, and it suggests that there are at least three different disease ecologies. The variation in spatio-temporal patterns of RRV indicates a complex ecology that is unlikely to be explained by a single dominant transmission route across these three groupings. Therefore, there is need to explore socio-economic and environmental determinants of RRV disease at the statistical local area (SLA) level. Spatial distribution analysis and multiple negative binomial regression models were employed to identify the socio-economic and environmental determinants of RRV disease at both the city and local (ie, SLA) levels. The results show that RRV activity was primarily concentrated in the northeast, northwest and southeast areas in Brisbane. The negative binomial regression models reveal that RRV incidence for the whole of the Brisbane area was significantly associated with Southern Oscillation Index (SOI) at a lag of 3 months (Relative Risk (RR): 1.12; 95% confidence interval (CI): 1.06 - 1.17), the proportion of people with lower levels of education (RR: 1.02; 95% CI: 1.01 - 1.03), the proportion of labour workers (RR: 0.97; 95% CI: 0.95 - 1.00) and vegetation density (RR: 1.02; 95% CI: 1.00 - 1.04). However, RRV incidence for high risk areas (ie, SLAs with higher incidence of RRV) was significantly associated with mosquito density (RR: 1.01; 95% CI: 1.00 - 1.01), SOI at a lag of 3 months (RR: 1.48; 95% CI: 1.23 - 1.78), human population density (RR: 3.77; 95% CI: 1.35 - 10.51), the proportion of indigenous population (RR: 0.56; 95% CI: 0.37 - 0.87) and the proportion of overseas visitors (RR: 0.57; 95% CI: 0.35 - 0.92). It is acknowledged that some of these risk factors, while statistically significant, are small in magnitude. However, given the high incidence of RRV, they may still be important in practice. The results of this study suggest that the spatial pattern of RRV disease in Brisbane is determined by a combination of ecological, socio-economic and environmental factors. The possibility of developing an epidemic forecasting system for RRV disease was explored using the multivariate Seasonal Auto-regressive Integrated Moving Average (SARIMA) technique. The results of this study suggest that climatic variability, particularly precipitation, may have played a significant role in the transmission of RRV disease in Brisbane. This finding cannot entirely be explained by confounding factors such as other socio-ecological conditions because they have been unlikely to change dramatically on a monthly time scale in this city over the past two decades. SARIMA models show that monthly precipitation at a lag 2 months (=0.004,p=0.031) was statistically significantly associated with RRV disease. It suggests that there may be 50 more cases a year for an increase of 100 mm precipitation on average in Brisbane. The predictive values in the model were generally consistent with actual values (root-mean-square error (RMSE): 1.96). Therefore, this model may have applications as a decision support tool in disease control and risk-management planning programs in Brisbane. The Polynomial distributed lag (PDL) time series regression models were performed to examine the associations between rainfall, mosquito density and the occurrence of RRV after adjusting for season and auto-correlation. The PDL model was used because rainfall and mosquito density can affect not merely RRV occurring in the same month, but in several subsequent months. The rationale for the use of the PDL technique is that it increases the precision of the estimates. We developed an epidemic forecasting model to predict incidence of RRV disease. The results show that 95% and 85% of the variation in the RRV disease was accounted for by the mosquito density and rainfall, respectively. The predictive values in the model were generally consistent with actual values (RMSE: 1.25). The model diagnosis reveals that the residuals were randomly distributed with no significant auto-correlation. The results of this study suggest that PDL models may be better than SARIMA models (R-square increased and RMSE decreased). The findings of this study may facilitate the development of early warning systems for the control and prevention of this widespread disease. Further analyses were conducted using classification trees to identify major mosquito species of Ross River virus (RRV) transmission and explore the threshold of mosquito density for RRV disease in Brisbane, Australia. The results show that Ochlerotatus vigilax (RR: 1.028; 95% CI: 1.001 - 1.057) and Culex annulirostris (RR: 1.013, 95% CI: 1.003 - 1.023) were significantly associated with RRV disease cycles at a lag of 1 month. The presence of RRV was associated with average monthly mosquito density of 72 Ochlerotatus vigilax and 52 Culex annulirostris per light trap. These results may also have applications as a decision support tool in disease control and risk management planning programs. As RRV has significant impact on population health, industry, and tourism, it is important to develop an epidemic forecast system for this disease. The results of this study show the disease surveillance data can be integrated with social, biological and environmental databases. These data can provide additional input into the development of epidemic forecasting models. These attempts may have significant implications in environmental health decision-making and practices, and may help health authorities determine public health priorities more wisely and use resources more effectively and efficiently.

The incidence of many arboviral diseases is largely associated with social and environmental conditions. Ross River virus (RRV) is the most prevalent arboviral disease in Australia. It has long been recognised that the transmission pattern of RRV is sensitive to socio-ecological factors including climate variation, population movement, mosquito-density and vegetation types. This study aimed to assess the relationships between socio-environmental variability and the transmission of RRV using spatio-temporal analytic methods. Computerised data files of daily RRV disease cases and daily climatic variables in Brisbane, Queensland during 1985-2001 were obtained from the Queensland Department of Health and the Australian Bureau of Meteorology, respectively. Available information on other socio-ecological factors was also collected from relevant government agencies as follows: 1) socio-demographic data from the Australia Bureau of Statistics; 2) information on vegetation (littoral wetlands, ephemeral wetlands, open freshwater, riparian vegetation, melaleuca open forests, wet eucalypt, open forests and other bushland) from Brisbane City Council; 3) tidal activities from the Queensland Department of Transport; and 4) mosquito-density from Brisbane City Council. Principal components analysis (PCA) was used as an exploratory technique for discovering spatial and temporal pattern of RRV distribution. The PCA results show that the first principal component accounted for approximately 57% of the information, which contained the four seasonal rates and loaded highest and positively for autumn. K-means cluster analysis indicates that the seasonality of RRV is characterised by three groups with high, medium and low incidence of disease, and it suggests that there are at least three different disease ecologies. The variation in spatio-temporal patterns of RRV indicates a complex ecology that is unlikely to be explained by a single dominant transmission route across these three groupings. Therefore, there is need to explore socio-economic and environmental determinants of RRV disease at the statistical local area (SLA) level. Spatial distribution analysis and multiple negative binomial regression models were employed to identify the socio-economic and environmental determinants of RRV disease at both the city and local (ie, SLA) levels. The results show that RRV activity was primarily concentrated in the northeast, northwest and southeast areas in Brisbane. The negative binomial regression models reveal that RRV incidence for the whole of the Brisbane area was significantly associated with Southern Oscillation Index (SOI) at a lag of 3 months (Relative Risk (RR): 1.12; 95% confidence interval (CI): 1.06 - 1.17), the proportion of people with lower levels of education (RR: 1.02; 95% CI: 1.01 - 1.03), the proportion of labour workers (RR: 0.97; 95% CI: 0.95 - 1.00) and vegetation density (RR: 1.02; 95% CI: 1.00 - 1.04). However, RRV incidence for high risk areas (ie, SLAs with higher incidence of RRV) was significantly associated with mosquito density (RR: 1.01; 95% CI: 1.00 - 1.01), SOI at a lag of 3 months (RR: 1.48; 95% CI: 1.23 - 1.78), human population density (RR: 3.77; 95% CI: 1.35 - 10.51), the proportion of indigenous population (RR: 0.56; 95% CI: 0.37 - 0.87) and the proportion of overseas visitors (RR: 0.57; 95% CI: 0.35 - 0.92). It is acknowledged that some of these risk factors, while statistically significant, are small in magnitude. However, given the high incidence of RRV, they may still be important in practice. The results of this study suggest that the spatial pattern of RRV disease in Brisbane is determined by a combination of ecological, socio-economic and environmental factors. The possibility of developing an epidemic forecasting system for RRV disease was explored using the multivariate Seasonal Auto-regressive Integrated Moving Average (SARIMA) technique. The results of this study suggest that climatic variability, particularly precipitation, may have played a significant role in the transmission of RRV disease in Brisbane. This finding cannot entirely be explained by confounding factors such as other socio-ecological conditions because they have been unlikely to change dramatically on a monthly time scale in this city over the past two decades. SARIMA models show that monthly precipitation at a lag 2 months (=0.004,p=0.031) was statistically significantly associated with RRV disease. It suggests that there may be 50 more cases a year for an increase of 100 mm precipitation on average in Brisbane. The predictive values in the model were generally consistent with actual values (root-mean-square error (RMSE): 1.96). Therefore, this model may have applications as a decision support tool in disease control and risk-management planning programs in Brisbane. The Polynomial distributed lag (PDL) time series regression models were performed to examine the associations between rainfall, mosquito density and the occurrence of RRV after adjusting for season and auto-correlation. The PDL model was used because rainfall and mosquito density can affect not merely RRV occurring in the same month, but in several subsequent months. The rationale for the use of the PDL technique is that it increases the precision of the estimates. We developed an epidemic forecasting model to predict incidence of RRV disease. The results show that 95% and 85% of the variation in the RRV disease was accounted for by the mosquito density and rainfall, respectively. The predictive values in the model were generally consistent with actual values (RMSE: 1.25). The model diagnosis reveals that the residuals were randomly distributed with no significant auto-correlation. The results of this study suggest that PDL models may be better than SARIMA models (R-square increased and RMSE decreased). The findings of this study may facilitate the development of early warning systems for the control and prevention of this widespread disease. Further analyses were conducted using classification trees to identify major mosquito species of Ross River virus (RRV) transmission and explore the threshold of mosquito density for RRV disease in Brisbane, Australia. The results show that Ochlerotatus vigilax (RR: 1.028; 95% CI: 1.001 - 1.057) and Culex annulirostris (RR: 1.013, 95% CI: 1.003 - 1.023) were significantly associated with RRV disease cycles at a lag of 1 month. The presence of RRV was associated with average monthly mosquito density of 72 Ochlerotatus vigilax and 52 Culex annulirostris per light trap. These results may also have applications as a decision support tool in disease control and risk management planning programs. As RRV has significant impact on population health, industry, and tourism, it is important to develop an epidemic forecast system for this disease. The results of this study show the disease surveillance data can be integrated with social, biological and environmental databases. These data can provide additional input into the development of epidemic forecasting models. These attempts may have significant implications in environmental health decision-making and practices, and may help health authorities determine public health priorities more wisely and use resources more effectively and efficiently.

Infectious diseases are a major public health burden causing millions of deaths every year. Government authorities need to be able to monitor disease incidence and evaluate their interventions for disease control. Monitoring the status of infectious diseases is one of the most challenging problems facing the public health sector, and epidemiological surveillance systems for infectious diseases, particularly notifiable diseases are essential. Despite initiatives to encourage reporting of infectious diseases, underreporting and poor surveillance are on-going challenges for many developing countries. Most surveillance systems in these settings use traditional paper-based methods, which are both inefficient and impractical. There is a need for alternative tools to strengthen infectious disease surveillance systems in resource-limited settings. The remarkable progress made in mobile computing technology has the potential to improve infectious disease surveillance systems. However, user experience in digital technologies and infrastructure needs to be given greater attention. My study investigated the use of mobile phone devices as surveillance tools in health information systems. A mobile phone-based surveillance system was developed and applied in Tanzania as an alternative approach to traditional paper-based systems. Using this system different factors that affect the usability of mobile phone-based systems were investigated, by examining the quality of surveillance data in the context of completeness, timeliness and costs. After two years of operation in twenty-eight districts in southern Tanzania, numerous factors were identified that affect user accuracy and speed of use of the mobile phone-based surveillance. These include user experience in digital technology, particularly mobile device ownership; digital technology literacy, such as access and use of SMS and user’s age. The mobile phone-based surveillance system was more accurate compared to the traditional paper-based system with greater data reporting, more complete data and timelier reporting. Initially the mobile phone-based surveillance system required more capital investment, although the running costs of paper-based surveillance were greater. The utility of the mobile phone-based surveillance in monitoring and evaluating large-scale rabies control interventions was examined and the data produced was used to analyse the impacts of interventions on reducing disease incidence. Significant relationships were detected between the incidence of reported bite injuries in the focal district the previous month and in neighbouring districts that month, with more injuries detected in mainland Tanzania than on the island of Pemba. The relationship between bite injuries and vaccination coverage was complicated, with some evidence that vaccination reduced bite incidence. However, more data and a better model are needed to fully understand the impact of vaccination on bite incidence. The system provided timely information on the implementation of control measures and incidence of bite injuries, vital for improving control efforts. Use of automated short text messages (SMS) as part of the mobile phone-based surveillance was assessed to determine whether they could improve patient’s adherence to treatment regimens. Patients who received SMS reminders had significantly better compliance than those who did not, with attendance improved by at least 10%. Use of SMS reminders has the potential to improve patients’ compliance in other treatment regimens that require repeat clinic visits or administration of medicines. This thesis documented how the use of mobile phone devices can be used to improve surveillance in resource-limited settings. The use of effective integrated surveillance system could empower major stakeholders concerned with public health problems by providing them with appropriate real-time information on disease incidence and control interventions. In the final chapter the challenges encountered and insights gained in the application of mobile computing in strengthening infectious diseases surveillance are discussed. Despite infrastructural challenges such as unreliable power and Internet, mobile computing technologies can improve patient care and authorities can be prompted in a timely manner about infectious disease outbreaks and of supply shortages. In conclusion, innovative tools that can strengthen and integrate human and animal surveillance can improve the control and prevention of infectious diseases. Mobile phones have great potential for this, and can be used to strengthen health information systems.

Les collisions de la route, une préoccupation mondiale en matière de santé publique et de socio-économie, coûtent la vie à 1,19 million de personnes chaque année. Le Liban, avec une population de moins de 6 millions d'habitants, enregistre chaque année près de 1 000 décès dus aux collisions de la route. Les défis sont exacerbés par la croissance démographique, la possession de véhicules au sein des foyers, la détérioration des infrastructures et les conflits régionaux, le tout compliqué par des données fragmentées. Le manque de recherches antérieures sur la sécurité routière aggrave la situation en entravant la formulation de politiques éclairées. En utilisant une approche multidimensionnelle, cette thèse entreprend une exploration exhaustive des défis de la sécurité routière, visant à combler les lacunes de la recherche et s'efforce de fournir une stratégie holistique pour améliorer la sécurité routière. Sa vision ultime s'étend à des routes plus sûres, à une réduction des décès et à l'évolution vers un système de transport durable. La thèse se déploie en deux axes principaux : le système de management de la sécurité routière et le comportement des conducteurs. Dans le premier axe, une analyse exhaustive de la situation actuelle du système de management de la sécurité routière au Liban a été réalisée. Cette analyse examine en profondeur les défis, ainsi que les complexités historiques, institutionnelles, organisationnelles et financières qui influencent le cadre actuel de la sécurité routière. La thèse évalue les mécanismes d'application du code routier, les affectations budgétaires et propose des recommandations fondées sur des données probantes pour améliorer à la fois la sécurité routière et la durabilité au Liban. Cette analyse exhaustive souligne l'urgence d'adopter une stratégie holistique couvrant un large éventail d'aspects liés à la sécurité routière, de l'application rigoureuse des lois à l'amélioration des infrastructures, en passant par l'allocation durable des financements et la collaboration internationale active. Reconnaissant que les données soient la pierre angulaire d'une gestion efficace de la sécurité routière, la thèse aborde brièvement le problème crucial de la fragmentation et de l'incomplétude des données sur les traumatismes liés à la circulation routière au Liban. Se penchant sur le facteur humain, le deuxième axe se concentre sur le comportement des conducteurs, élément essentiel de la sécurité routière qui a manqué d'outils validés dans le contexte libanais. La thèse comporte l'adaptation transculturelle de quatre échelles de mesure (le locus de contrôle du trafic, le questionnaire sur le comportement des conducteurs, l’échelle de la colère au volant et l'inventaire des compétences des conducteurs) pour comprendre et atténuer les comportements dangereux des conducteurs. Ce processus complexe implique la traduction, l'adaptation transculturelle, une validation rigoureuse et l'implication de divers groupes de conducteurs libanais dans la collecte de données. Au-delà de la validation, elle explore l'association entre ce comportement et les collisions de la route, avec le potentiel d'influencer les politiques de sécurité routière et de promouvoir de bonnes pratiques de conduite. Cette thèse sert alors de plan directeur transformateur, visant à forger des routes plus sûres, à réduire les collisions et à améliorer la qualité de vie au Liban. Il s'agit d'un appel sans équivoque à répondre à l'urgence d'agir et à promouvoir des solutions fondées sur des données probantes
Road traffic collisions, a global public health and socio-economic concern, claim 1.19 million lives annually worldwide. Lebanon, with a population under 6 million, contends with nearly 1,000 road fatalities each year. Challenges are amplified by population growth, increased household vehicle ownership, deteriorating infrastructure, and the enduring impact of regional conflicts, all further obscured by fragmented data. The dearth of prior research on Lebanese road safety compounds the predicament, impeding informed policy formulation and precise solutions. The thesis unfolds along two main axes: road safety management and driver behaviour. The first axis provides a comprehensive analysis of the current state of road safety management in Lebanon. This analysis delves deeply into the challenges, as well as the historical, institutional, organizational, and financial complexities that shape the current framework for road safety. The thesis evaluates existing policies, enforcement mechanisms, budget allocations, and proposes evidence-based recommendations to improve both road safety and sustainability in Lebanon. This thorough analysis highlights the urgent need for a holistic strategy that covers a wide range of road safety aspects, from strict law enforcement to infrastructure improvements, sustainable funding allocation, and active international collaboration. Recognizing that data is the cornerstone of effective road safety management, the thesis addresses briefly the critical issue of fragmented and incomplete data on road traffic injuries in Lebanon. Focusing on the human factor, the second axis concentrates on driver behaviour, a key element of road safety that has lacked validated tools in the Lebanese context. The thesis advocates for the cross-cultural adaptation of four measurement scales (Traffic Locus of Control, Driver Behaviour Questionnaire, Driving Anger Scale, and Driver Skills Inventory) to understand and mitigate risky driver behaviours. This complex process involves translation, cross-cultural adaptation, rigorous validation, and the involvement of various groups of Lebanese drivers in data collection. Beyond validation, it explores the relationship between driver behaviour and road collisions, with the potential to influence road safety policies and promote good driving practices. This thesis thus serves as a transformative roadmap, aiming to create safer roads, reduce collisions, and improve the quality of life in Lebanon. It is a clarion call to address the urgent need for action and the pursuit of data-driven solutions

碩士
亞洲大學
資訊工程學系碩士班
96
In recent year the growth of smart digital surveillance system is boom. Except for traditional surveillance functions, a new multi-camera surveillance system has functions of automatic detection, tracing moving objects, identifying the moving object and behavior analysis. New functions focus on anomaly detection and object identification but are lack of the relations between each camera and the functions are also lack of the capability of querying about history surveillance pictures. Therefore the paper proposes an environmental surveillance system which can relate each camera and integrate functions of object tracing, object identifying, recording and querying object features. The proposed multi-camera surveillance system is able to execute real time security protection and emergency management. The paper is based on computer vision to use several cameras to construct an environmental surveillance system. The cameras can be deployed in any kind of place, for example, to the security management of place with bad social order, to residence and to office buildings. The experiment has proved that through the connections between cameras build by the system in a real scene, the system is able to record object features sufficiently, to trace object in real time, to reduce the time for querying history records and to increase the capability of emergency management.

碩士
國立交通大學
資訊學院資訊科技產業專班
98
In conventional surveillance systems, multiple screens are often required for displaying video from multiple cameras and may cause the difficulty of operators to keep track of targets due to the lack of special relationship among the screens. In this thesis, we develop an effective surveillance system with the 3-D environment model that integrates multiple scenes into one single comprehensive view. The system does not require accurate camera calibration and environment model construction with advanced equipments, and can provide a multiscale operating view for showing the status of the surveillance area. To integrate the monitored area with camera views, the 3-D environment is first manually constructed by planar patch modeling. To map video contents to the corresponding areas of the 3-D model, different homography transformations are estimated for every pairs of image regions in the video contents and corresponding areas in the 3-D model. The planar patches in the 3-D model are automatically divided into different sizes and numbers of the smaller patches are determined by the intensity differences between the image polygons obtained from the homography transformation and texture mapping. Lookup tables are built beforehand for accelerating the coordinate mapping. In monitored scenes, 3-D objects including pedestrians, are projected to the imaging sensor planes through perspective transformation. Therefore, all 3-D objects will appear flattened on the planes in the 3-D model after texture mapping. To overcome this problem, we use billboarding method to model the 3-D moving objects. First, each foreground object extracted from the scene by background modeling is mapped to a billboard, and then vertically aligned to the ground plane. In this way we can adjust the direction of foreground objects according to the viewing direction. The proposed system can provide operators the situational awareness of the monitored site, including activities of the tracking targets through a comprehensive 3-D view.

碩士
長榮大學
資訊管理研究所
97
The current socio-economic recession, the rise in unemployment, frequent criminal incidents, resulting in disturbing people''s minds,many people have insecurity.Have been stolen around the events, people and property of great importance to its attention, more, installation of anti-theft measures at home. Employment security company to patrol the community and the installation of security cameras, but cameras have a corner, afraid of the employment protection embezzlement. 　　This study has proposed a face recognition function of the alarm system. System to screen capture on the environment, when the images are inconsistent with the background map, we will ask people who come in Face Verification, not in time to verify or not validated that this person will be found is unknown, the system will inform the management , but through the authentication system can be shut down. 　　Video monitoring system in the traditional study, has made many important theoretical and technical. In this study, in addition to a variety of methods to implement monitoring systems, will focus on the following points: First, use the context of the registration law, registration of a fixed background, in order to filter out some pictures of the invasion; Second, using a simple method of face recognition , in a small sample set (under the assumption that less than 10 people), will identify ways to increase; three, combined with real-time E-mail with the voice of the implementation of the alarm system

碩士
淡江大學
電機工程學系碩士班
100
Due to the problem of extremely low birth rate,our society is turning rapidly to be an aging one. “An infant at home is a treasure you own” This shows vividly the phenomenon that will exist in the future. Each infant is a precious stone of his/her parents. However, the research on infant-care is insufficient. We have heard some unfortunate news of infants in recent years. In view of this, we develop a system for infant-care. No sensor device is attached to the infant. And the infant-care system we build adopt a non-contact approach for monitoring and pre-warning. Through the system, the parents can know well the situation of the infant at anytime. In this paper, the environment under consideration is in house. The parents can put the infant in the living room, and set up the camera to start the proposed system of surveillance. The infant’s activity can be shot by the camera. The parents can set a safety-zone for the infant via the GUI of the proposed system. When the system detects that the infant is close to the window or not in the range of surveillance, it will send a warning ringtone to notify the parents. Briefly, the system consists of two parts. First, the camera produces several images per second. The proposed system processes the captured images and tracks the infant’s position by Kalman Filter. Second, it is the platform Android mobile phone. The pre-warning notice is realized through the smart phone application. When the Kalman Filter predicts that the infant will crawl out of the defaulted safety-zone, it can activate the warning ringtone of mobile phone to notify the parents. We design a GUI interface by Matlab, which enables the user to operate and monitor. In addition, we use the software of Eclipse to create an application of Android mobile phone, which can synchronously operate and monitor from mobile phone. The proposed system integrates all the technologies for infant-homecare. The experimental results show the effectiveness of the proposed system.

Background: One Health is an interdisciplinary approach to zoonotic diseases, which encourages structured collaboration and coordination between human, animal and ecological sectors. This interdisciplinary approach could be applied to zoonotic disease surveillance in Australia. Addressing zoonoses at the interface of human, animal and environmental health is considered to be the most holistic approach to zoonotic disease control. Seventy five percent of emerging infectious diseases are of animal origin, so an approach that links the health of humans, animals and the environment could provide an earlier opportunity for disease detection and therefore may help to reduce the burden of zoonotic diseases. Currently in Australia, human, animal and environmental health are managed by separate sectors with limited communication, acting as a barrier to effective and timely zoonotic responses. This study aims to explore how professionals in the field of human, animal and ecological health perceive a One Health approach to zoonotic disease surveillance, aiming to identify what the challenges are to the implementation of an integrated surveillance system in Australia. Methods and results: Using a qualitative research method, ten semi-structured interviews were conducted with experts in the areas of human, animal and ecological health in order to gain an understanding of professional opinions regarding the challenges of implementing One Health surveillance in Australia. A thematic analysis of the data was undertaken to identify recurring themes. Findings showed that the absence of a clear definition of One Health acts as a barrier to collaboration, as well as siloed approaches by different sectors restricting the ability for professionals to work collaboratively across disciplines. Understanding disease transmission as a whole, as well as understanding the role of the environment on human and animal health were considered by participants to be vital requirements for a One Health approach to be successful. Additionally, political will was considered by participants to be an essential requirement for the integration of government systems. Conclusions: This study demonstrates that for a One Health approach to be implemented in an Australian setting, those working in the fields of human, animal and ecological health must come together to agree on a ‘One Health’ definition. Restructuring of the traditional silos, which currently restrict intersectoral collaboration, could result in an improved and collective approach to zoonotic disease surveillance. This could be achieved through the establishment of a formal governance body. Regular communication may provide an avenue for interdisciplinary approaches, and could assist in overcoming the longstanding barriers of privacy and distrust between sectors. Further, developing interdisciplinary public health training in medical, environmental and veterinary degrees may encourage cross-disciplinary collaboration. Finally, illustrating the economic benefit of faster zoonotic detection will likely attract the attention of politicians, who could assist in implementing a formal and structured One Health approach in Australia.
Thesis (BHlthSc(Hons)) -- University of Adelaide, School of Public Health, 2016

碩士
臺灣大學
電機工程學研究所
98
In recent years, much research has been focused on functionality and feasibility of multi-camera surveillance system. This study describes two types of surveillance scenarios. One is the surveillance of public monitoring and the other is the surveillance of indoor environment with numerous stations. In the type of surveillance of public monitoring in wide area, the proposed architecture is designed. On the aspect of surveillance system, the wider coverage guarantees the security of area. However, it is difficult to gather detailed information using the wide field-of-view (FOV) sensor due to its limitation in resolutions. In order to maintain the desired image resolution and still have a wide FOV, this requires the use of active camera. Thus, the current architecture design combined fixed global-view camera and active focused-view camera to make use of their advantage, respectively. Furthermore, the methods to achieve multi-target object detection and tracking are proposed. In order to maintain the identity of a moving object, the trajectory of the center of mass (TCM) is proposed to accomplish this task of labeling. To coordinate two different sensors, the model of coordinate transformation is derived. Without the act of resource assignment, system redundancy is increased during the surveillance process. Hence, the system aims to reduce this redundancy by applying the cooperative strategy. In the scenario of indoor environment with numerous stations (e.g., classroom, assemble line in factory, office), multiple observation points are required for visual sensors. The method of monitoring multiple points is proposed to improve the correctness of observed points for further information transmission with other global-view cameras. The performance of motion detection algorithm is occasionally poor due to its fixed background updating rate. The problem with the fixed low background updating rate is that static object is not updated as the background since the transient time is short. However, with the fixed high background updating rate, the result of updating moving objects as the background is not desired. To improve the correctness of detecting result, motion detection with the adaptive background updating (ABU) is proposed. Finally, the experimental results of different scenarios are shown in this study.

碩士
臺北醫學大學
醫學資訊研究所
96
Syndromic surveillance is the utillization of data that large amount of cases with prodromal phase symptoms and performs spatial-temporal clustering analysis through information technology for rapidly detecting disease outbreaks with visualization and aberration detection theory so that a further epidemiology investigation and disease control procedure could be taken in its very early time.In recent years, many countries have adopted syndromic surveillance systems as the frontline defense against emerging infectious diseases or bioterrorism attacks. Although Taiwan’s Centers for Disease Control, R.O.C.(Taiwan CDC) has established the system for several years, the disease control staff and investigators in local health agencies and hospitals are lack of experience in using syndromic surveillance systems for prevention of infectious disease outbreaks. Therefore, this study focuses on: (1) the establishment of an integrated syndromic surveillance system that can automatically collect data from hospital emergency departments (ED) in Taipei City in a timely and flexible fashion, (2) development of effective algorithms for early detection of disease outbreaks using data of routinely collected chief complaints or ICD-9 CM codes, and (3) provision of friendly interfaces for the presentation of surveillance data to provide information for healthcare workers and decision-makers at different levels of position in hospitals in order to enhance the capability in the detection and prevention of disease outbreaks. The study selected five hospitals located in different geographical areas of Taipei City. The system has established and collected data of patient visiting ED (totally about 500,000 visits) during the period of January 1, 2005 and June 30, 2008. There are 8 syndrome groups for routine surveillance. In addition, the system can dynamically define new groups based on the trends of disease occurrence. Until now 5 dynamically defined syndrome groups have been performed in this system as an extra targeted surveillance. The historical limit method with the short-term and long-term baseline data is used for analysis in aberration detection, and the alerts are delivered over to persons concerned via email once it is detected. Meanwhile, the analyzed surveillance data, which is accessible through pages on website, can be used to compare the parallel trends of disease occurrence in different years. Furthermore, they can also be linked to a geographic information system to view the dynamic changes in temporal and spatial patterns of disease occurrence, the occurrence scale of clustering cases, and the development of trends for any interested syndromes. Since our system is mainly used for surveillance of locally important disease and covered only Taipei area in geographic, in the future, we will make our effort to set our disease defense strategies by grading them into different stages in the processes of preparation and response, so that it can work as a whole with those conducted by the central government, to get better system performance.

The study was conducted in the Greater Kruger National Park Complex (GKNP), which consists of the Kruger National Park (KNP) and adjacent private game reserves and focuses primarily on the following objectives: • To monitor and evaluate the standard of the existing disease surveillance programmes for the following diseases, Foot and Mouth, Anthrax, Tuberculosis, Brucellosis and Rabies, within the Kruger National Park and adjacent private game reserves by evaluating the level of competency and knowledge in field rangers, field guides and trails rangers with regard to these specific diseases. It can be stated that important differences exist between disease surveillance techniques used for domestic animals and those used for wildlife (Bengis, R.G., Kock, R.A., & Fischer, J., 2002). According to Morner, T., Obendorf, D.L., Artios, M., & Woodford, M.H., 2002, it is more difficult to monitor diseases in wildlife than in domestic animals because wild animals are not constrained by boundaries and can roam over large. distances. For significant diseases in wildlife, an active surveillance programme may be the preferred approach with the aim to collect a certain number of samples from a target population (live or dead animals) to determine the point prevalence of certain pathogens. Active veterinary participation is essential in protected area management, with emphasis on training of technicians, rangers and field biologists with regard to specific diseases and their clinical signs, surveillance and sampling techniques, data collection, and reporting. For the purpose of this study, data collection was conducted by means of a questionnaire drawn up according to the related critical points as described in the Dufour grid (Dufour, 1998). The results of this study clearly showed a need to address certain important aspects regarding a wildlife disease programme within the GKNP. A more efficient wildlife disease surveillance programme, which included more specific and “hands-on” trained staff, would definitely ensure a better early warning system which would detect new or emerging disease outbreaks.
M. Tech. (Nature Conservation))