Dissertations / Theses on the topic 'Visual and time-based digitalthesis'
To see the other types of publications on this topic, follow the link: Visual and time-based digitalthesis.
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 42 dissertations / theses for your research on the topic 'Visual and time-based digitalthesis.'
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
Tolmie, Julie, and julie tolmie@techbc ca. "Visualisation, navigation and mathematical perception: a visual notation for rational numbers mod1." The Australian National University. School of Mathematical Sciences, 2000. http://thesis.anu.edu.au./public/adt-ANU20020313.101505.
Full textAbstract:
There are three main results in this dissertation.
The first result is the construction of an abstract visual space for rational
numbers mod1, based on the visual primitives, colour, and rational radial
direction. Mathematics is performed in this visual notation by defining
increasingly refined visual objects from these primitives. In particular,
the existence of the Farey tree enumeration of rational numbers mod1
is identified in the texture of a two-dimensional animation.
¶
The second result is a new enumeration of the rational numbers mod1,
obtained, and expressed, in abstract visual space, as the visual object
coset waves of coset fans on the torus. Its geometry is shown to encode
a countably infinite tree structure, whose branches are cosets, nZ+m,
where n, m (and k) are integers. These cosets are in geometrical 1-1
correspondence with sequences kn+m, (of denominators) of rational
numbers, and with visual subobjects of the torus called coset fans.
¶
The third result is an enumeration in time of the visual hierarchy of the
discrete buds of the Mandelbrot boundary by coset waves of coset fans.
It is constructed by embedding the circular Farey tree geometrically into
the empty internal region of the Mandelbrot set. In particular, coset fans
attached to points of the (internal) binary tree index countably infinite
sequences of buds on the (external) Mandelbrot boundary.
2
Blagrove, Elisabeth. "Time-based visual selection with emotional faces." Thesis, University of Warwick, 2009. http://wrap.warwick.ac.uk/3623/.
Full textAbstract:
The biological and behavioural importance of the face has led to the proposition of several mechanisms dedicated to highly efficient specialized processing (e.g., M.H. Johnston, 2005). This is reflected in the attentional properties attributed to facial stimuli, especially when they contain affective information (e.g., R. Palermo & G. Rhodes, 2007). This thesis examines those attentional properties via a modified version of the visual search paradigm (i.e. the preview search task; D.G. Watson & G.W. Humphreys, 1997), which proposes that observers can intentionally suppress items seen prior to a full search array, for effective search performance (i.e. the preview benefit; D.G.Watson & G.W. Humphreys, 1997, 1998). The findings from this thesis show that it is possible to deprioritize previewed facial stimuli from search, although only a partial preview benefit was shown. Emotional valence of previewed faces had little impact on this effect, even when preview duration was extended from 1000-3000ms. However, when duration was reduced to 250-750 ms, negatively valenced faces were more difficult to suppress than positively valenced faces. In addition, when previewed faces changed expression concurrently with the onset of the full search array, the preview benefit was abolished, irrespective of the direction of the expression change (i.e. neutral to positive, or neutral to negative). A search advantage for negative face targets was demonstrated throughout all of the investigations in this thesis. These findings are consistent with previous work establishing preferential detection of, and selectively impaired disengagement from, negative faces (e.g., J.D. Eastwood, D. Smilek, & P.M. Merikle, 2001; E.Fox, R. Russo, R.J.Bowles, & K. Dutton, 2001). However, they also suggest the sensitivity of the visual marking mechanism to ecological considerations (such as the nature of the stimulus), and the overall relevance of emotional face stimuli to the visual system.
3
Zupan, Zorana. "Control and development of time-based visual selection." Thesis, University of Warwick, 2015. http://wrap.warwick.ac.uk/85423/.
Full textAbstract:
Attention plays an integral role in healthy cognitive functioning, and failures of attention can lead to unfavourable and dangerous consequences. As such, comprehending the nature of attentional mechanisms is of fundamental theoretical and practical importance. One way in which humans can attentionally prioritise new information is through top-down inhibition of old distractors, known as the preview benefit (Watson & Humphreys, 1997). In the preview benefit, time is used to efficiently guide visual selection in space. Given that this ability is based on limited resources, its deployment in everyday life may be hindered by a multitude of factors. This thesis will explore the endogenous and exogenous factors that can facilitate or constrain the preview benefit, and determine its developmental trajectory. Understanding the nature of this mechanism (endogenous and exogenous factors) in adults can elucidate the contexts in which visual selection can efficiently filter old distractors. In turn, a developmental perspective can unravel the hidden aspects of this ability and inform when children are endowed to use temporal information for efficient attentional selection. Chapter 1 introduces the theoretical problems and topics of attentional research in adults and children. Chapter 2 addresses the question of endogenous control of top-down inhibition in time-based visual selection – when can top-down inhibition be controlled by the observer? Chapter 3 examines the exogenous influence of complex stimuli on time-based visual selection. Chapters 4 and 5 focus on the development of time-based visual selection for stationary and moving stimuli, respectively, in children aged 6 to 12 years. These chapters also examine the relative association of the efficiency of the preview benefit with the development of executive functions across different age-groups. Overall, the findings suggest that there exist remarkable endogenous and exogenous constraints in how time guides selection. This may account for why in certain contexts, attentional selection can fail to be efficient. Moreover, time-based visual selection shows striking quantitative and qualitative changes over developmental time, and most importantly, children have a long developmental trajectory in learning to ignore moving items. Unlike children, adults’ time-based visual selection is coupled with individual differences in executive functions, highlighting an acquired functional connection. The findings are discussed in terms of their theoretical implications for time-based visual selection, the development of children’s attentional control for distractors, and impact routes for educational and clinical practice, and policy makers.
4
Whig, Alok. "Stream processor based real-time visual tracking using appearance based approach." [Gainesville, Fla.] : University of Florida, 2009. http://purl.fcla.edu/fcla/etd/UFE0025166.
Full text
5
Lai, Sui-man, and 賴萃文. "Design of a time-encoded visual stimulation method for brain computer interface based on chromatic transient visual evoked potentials." Thesis, The University of Hong Kong (Pokfulam, Hong Kong), 2009. http://hub.hku.hk/bib/B43085829.
Full text
6
Lai, Sui-man. "Design of a time-encoded visual stimulation method for brain computer interface based on chromatic transient visual evoked potentials." Click to view the E-thesis via HKUTO, 2009. http://sunzi.lib.hku.hk/hkuto/record/B43085829.
Full text
7
Nguyen, Anthony Giang. "Rate control based on human visual perception characteristics and its application in encoding time-lapse video /." Thesis, Connect to this title online; UW restricted, 2002. http://hdl.handle.net/1773/5921.
Full text
8
Diab, Hassan Bahaeddine. "Microprocessor based real-time modelling in relation to the visual systems of an aircraft flight-training simulator." Thesis, University of Bath, 1985. https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.353694.
Full textAbstract:
The purpose of this thesis is to show the application of modern digital techniques to produce a high speed computational system that has the capability of producing a faster than real-time model for the visual systems associated with flight simulator systems. A distributed/multiprocessing system was chosen due to the increased reliability, flexibility and throughput of the computational process. The system also has the advantage of providing increased modularity and ease of expansion. This research is firstly concerned with the development of a high resolution graphics system (providing a 1024 by 1024 point screen resolution), and the techniques for the production of moving visual scenes observed by a pilot when flying over a region of terrain in a flight simulator. In particular, this work is concerned with the real-time simulation of the aerodynamics of the aircraft and considers the problems of designing and accessing a data base which represents a model of the terrain. There are three major design considerations for the real-time CGI (Computer Generated Imagery) system of the type required. The first is to ensure that all processing for each picture frame is performed within a specified time limit. Secondly, the picture must have a realistic appearance and requires realism functions to be performed in real-time. The third design consideration is the cost.
9
Dippold, Amanda. "Vision-Based Obstacle Avoidance for Multiple Vehicles Performing Time-Critical Missions." Diss., Virginia Tech, 2009. http://hdl.handle.net/10919/27830.
Full textAbstract:
This dissertation discusses vision-based static obstacle avoidance for a fleet of
nonholonomic robots tasked to arrive at a final destination simultaneously. Path generation
for each vehicle is computed using a single polynomial function that incorporates the vehicle
constraints on velocity and acceleration and satisfies boundary conditions by construction.
Furthermore, the arrival criterion and a preliminary obstacle avoidance scheme is incorporated
into the path generation. Each robot is equipped with an inertial measurement unit
that provides measurements of the vehicleâ s position and velocity, and a monocular camera
that detects obstacles. The obstacle avoidance algorithm deforms the vehicleâ s original path
around at most one obstacle per vehicle in a direction that minimizes an obstacle avoidance
potential function. Deconfliction of the vehicles during obstacle avoidance is achieved by
imposing a separation condition at the path generation level. Two estimation schemes are
applied to estimate the unknown obstacle parameters. The first is an existing method known
in the literature as Identifier-Based Observer and the second is a recently-developed fast estimator.
It is shown that the performance of the fast estimator and its effect on the obstacle
avoidance algorithm can be arbitrarily improved by the appropriate choice of parameters as
compared to the Identifier-Based Observer method. Coordination in time of all vehicles is
completed in an outer loop which adjusts the desired velocity profile of each vehicle in order
to meet the simultaneous arrival constraints. Simulation results illustrate the theoretical
findings.Ph. D.
10
Haegele, Justin. "The Effects of a Theory Based Physical Education Intervention on the Leisure-Time Physical Activity of Adolescents with Visual Impairments." The Ohio State University, 2015. http://rave.ohiolink.edu/etdc/view?acc_num=osu1429085562.
Full text
11
Tong, Xin. "Interactive Visual Clutter Management in Scientific Visualization." The Ohio State University, 2016. http://rave.ohiolink.edu/etdc/view?acc_num=osu1471612150.
Full text
12
Yu, Kuan Tao. "Development of a PC software package using windows 95 and visual C++ to evaluate traffic safety improvements based upon accidents per unit time." Ohio University / OhioLINK, 1996. http://rave.ohiolink.edu/etdc/view?acc_num=ohiou1177617333.
Full text
13
Sharif, Abbass. "Visual Data Mining Techniques for Functional Actigraphy Data: An Object-Oriented Approach in R." DigitalCommons@USU, 2012. https://digitalcommons.usu.edu/etd/1394.
Full textAbstract:
Actigraphy, a technology for measuring a subject's overall activity level almost continuously over time, has gained a lot of momentum over the last few years. An actigraph, a watch-like device that can be attached to the wrist or ankle of a subject, uses an accelerometer to measure human movement every minute or even every 15 seconds. Actigraphy data is often treated as functional data. In this dissertation, we discuss what has been done regarding the visualization of actigraphy data, and then we will explain the three main goals we achieved: (i) develop new multivariate visualization techniques for actigraphy data; (ii) integrate the new and current visualization tools into an R package using object-oriented model design; and (iii) develop an adaptive user-friendly web interface for actigraphy software.
14
Mercado, Angel, Hervin Vargas, Edgardo Carvallo, and Carlos Raymundo. "Proposal to optimize the flow of preparation and delivery of vehicles to dealers based on the Lean methodology." Latin American and Caribbean Consortium of Engineering Institutions, 2019. http://hdl.handle.net/10757/656262.
Full textAbstract:
Currently, there is a monthly breach with its vehicle delivery policy. This policy refers to said company, it refers to a minimum of 92% of units towards the points of sale. However, in the last quarter of 2017 and the semester of 2018, it has only been met from 82% to 90%. These are 5% -10% that have not come to meet in each month, but also directly to the profitability of the company. On the one hand, storage costs are increased by each car that does not arrive or dissatisfaction is generated by the customer, this is the problem of 3 main causes, which is the installation time of radios, installation of alarms and more vehicles that are not served in the damage assessment area. As a root cause, we have the first, since, on the one hand, a bottleneck is generated by the lack of car attention and, on the other hand, represents 20% of the monthly sales of the case study, For what needs to be done to reduce this bottleneck and increase the capacity of the vehicles in this area, the damage is detailed, two lines of evaluation are carried out. This type of triage, we ensure a standardized work, since there was a group of vehicles waiting with different types of damage in the same operating line. Also, the delivery of vehicles on time is improved to 95% and the times of the areas of installation of radios, installation of alarms and damage assessment are reduced by 42.85%, 51.42% and 50% respectively.
15
Morosanu, Roxana. "Presents of the Midlands : domestic time, ordinary agency and family life in an English town." Thesis, Loughborough University, 2014. https://dspace.lboro.ac.uk/2134/16243.
Full textAbstract:
Focusing on the everyday lives of middle-class English families in a medium size town situated in the Midlands, this doctoral thesis contributes to anthropological debates on the topics of human agency, time, domesticity, mothering, and kinship. Organized upon the idea that cultural models of time are inextricably linked to understandings of agency (Greenhouse 1996), the thesis links Moore s (2011) post-vitalist theoretical framework and the work of Foucault (1990, 2000) on ethical practices, with Gershon s (2011) critique of neoliberal agency . The concept of ordinary agency is proposed for situating everyday actions as significant actions that contribute to social transformation. Three cultural models of time are identified spontaneity, anticipation and family time and the types of ordinary agencies that they engage are described in three dedicated chapters. The first chapter discusses the theoretical framework of the thesis. The second chapter addresses methodological issues, and discusses the methods that the author developed during her ethnographic fieldwork for looking at people s relationships with time. The third chapter addresses the time mode of spontaneity, presenting ethnographic examples of digital media use at home, and introducing theoretical tools for situating the forms of agency engendered by spontaneity. The fourth chapter looks at the time mode of anticipation in relation to mothering, motherhood and care. This chapter is accompanied by a video component, titled Mum s Cup and situated in the appendix of the thesis. Based on material that the participants filmed in solitude, for a self-interviewing with video task, Mum s Cup is a visual point of departure for theorising the Mother-Multiple ontological position that is described in chapter IV. Alongside providing a visual ethnographic lever for endorsing a theoretical concept, the video project also reflects on the relationship between the researcher and the participants, a relationship that, for various reasons (some related to length limitations), is not fully described in the textual corpus of the thesis. Discussing two types of domestic sociality, the fifth chapter looks at family time and at the forms of agency engendered by the idea and by the experience of having a family-style lifestyle (Strathern 1992), and it draws on, and contributes to, bodies of literature on English kinship. The last chapter addresses the context of the research which is an interdisciplinary project looking at domestic energy consumption ; it situates the position of the author in relation to the domestic sustainability agenda and to debates on interdisciplinarity, and it formulates ideas about possible applications that the anthropological knowledge gained by the author through her research could have in relation to the context that originally framed and facilitated the research.
16
Lieberknecht, Sebastian Verfasser], Nassir [Akademischer Betreuer] [Navab, and Gerhard [Akademischer Betreuer] Reitmayr. "Contributions to Real-time Visual Tracking and Live Environment Modeling for Augmented Reality Applications : Design and Implementation of New Methods and Evaluations on Real Image-based Ground Truth Data / Sebastian Lieberknecht. Gutachter: Nassir Navab ; Gerhard Reitmayr. Betreuer: Nassir Navab." München : Universitätsbibliothek der TU München, 2013. http://d-nb.info/1031280359/34.
Full text
17
Ma, Zhenyu. "Semi-synchronous video for Deaf Telephony with an adapted synchronous codec." Thesis, University of the Western Cape, 2009. http://etd.uwc.ac.za/index.php?module=etd&action=viewtitle&id=gen8Srv25Nme4_2950_1370593938.
Full textAbstract:
Communication tools such as text-based instant messaging, voice and video relay services, real-time video chat and mobile SMS and MMS have successfully been used among Deaf people. Several years of field research with a local Deaf community revealed that disadvantaged South African Deaf
people preferred to communicate with both Deaf and hearing peers in South African Sign Language as opposed to text. Synchronous video chat and video
relay services provided such opportunities. Both types of services are commonly available in developed regions, but not in developing countries like South
Africa. This thesis reports on a workaround approach to design and develop an asynchronous video communication tool that adapted synchronous video
codecs to store-and-forward video delivery. This novel asynchronous video tool provided high quality South African Sign Language video chat at the
expense of some additional latency. Synchronous video codec adaptation consisted of comparing codecs, and choosing one to optimise in order to
minimise latency and preserve video quality. Traditional quality of service metrics only addressed real-time video quality and related services. There was no
uch standard for asynchronous video communication. Therefore, we also enhanced traditional objective video quality metrics with subjective
assessment metrics conducted with the local Deaf community.
18
Liang, Shin-Hwei, and 梁欣蕙. "Feature-Based Visual Speech Recognition Using Time-Delay Neural Network." Thesis, 1997. http://ndltd.ncl.edu.tw/handle/10921510042799247430.
Full textAbstract:
碩士國立交通大學
控制工程系
85
An automatic mouth feature detection and mouth motion recognition technique for visual speech recognition is proposed in this thesis. This technique consists of three stages : human mouth detection and extraction, mouth feature detection, and neural network learning. In the mouth detection stage, the first step is to find the locations of human faces without any constraints on the users for the consideration of practicability. Hough transform is used here for determining the candidate face locations under complex environments. We simplify it to a three-dimensional search and redefine the searching region using the symmetry property of human beings. Then, a Mouth Detection Algorithm (MDA) is proposed to verify the mouth location and the next three procedures are normalization, adjustment, and template matching for the candidate mouth images. After these processes only one mouth image is treated as the winner among the candidate mouth images. In the mouth feature detection stage, one procedure searches the mouth corners and a refined Mouth Feature Searching Algorithm (MFSA) is used to reconnoiter the four points on two lips. These four points play an important role in our system since two parabolas can be approximated using the mouth corners and these points. Finally, a precise mouth model is established after calculating two parabolas and selecting eleven features from the mouth model as the input patterns for the classifier. In the last stage, a TDNN is used as our classifier due to the tolerance of time shifting property. We have done many experiments to decide which kinds of features are crucial and sufficient enough in the lip- reading system. The off-line recognition rate can achieve 90% speaker dependently in our experiment. Two other methods are compared with our system and we find that our method can reach better performance than other two methods with the less memory space and training time. Finally, we generalize our system to a six speakers system to verify the robustness of our method. The experimental result shows the stability and practicability of the proposed approach.
19
Liu, Geng-Hua, and 劉耿華. "Real-time Humanoid Visual System based on Region Segmentation for obstacle determination." Thesis, 2011. http://ndltd.ncl.edu.tw/handle/87404192429983951785.
Full textAbstract:
碩士國立臺灣大學
機械工程學研究所
99
This thesis proposes one real-time visual system completely, and the suitable control methods are developed for the especial humanoid eye mechanism by our self. This eye mechanism has five degree of freedoms to control the horizontal and vertical rotations of two eyeballs and opening of two eyelids. PMMA is chosen as the material for convenience of CNC machining and its lightness. Those used motors and control boards are also mini and light. The analyzed images are inputted from mini webcams, and then the visual program use region-based corresponding result between images with different viewpoints and disparities of those corresponded regions to calculate the spatial distances between the eye mechanism and those objects represented by corresponded region pairs. For the balance between efficacy of region separation and computing efficiency to satisfy real-time operation, this research uses the algorithm of k-mean classification as the basis to classify image pixels by color statistics and also improves the color space to let it more uniform in description of chromatism for better efficacy. The accelerative program mechanism for regions extraction is also designed to let the passing times of each region boundary decrease to only one in actual searching stage. Those centroid pairs from corresponded regions are considered that the basis of distance determination. The errors of calculated distance are affected by parameters calibration of image sensors and precision of eyeballs’ rotational angles control. Therefore, the eye mechanism is also installed with angle gauges and analog-to-digital board to achieve real-time feedback of horizontal rotations. The revision of vertical rotation control by software is designed because the difficulty of installation and applying of the extra equation from coordinate transformation. To meet the requirement of real-time operation finally, this research also involves the help by Nvidia graphic cards and CUDA toolkits. After suitable parallelization of original serial program and operation on GPU of graphic cards, the efficiency of visual system obtains 5 to 10 times improvement by the higher ability in general computation of current GPU based system. Finally, the operation frequency of this visual system achieves more than 30 Hz which meets the general definition of real-time operation.
20
Lin, Shu-Wei, and 林書緯. "Real-Time Simultaneous Image Stitching and Localization Based on Visual Content Maximization." Thesis, 2013. http://ndltd.ncl.edu.tw/handle/bh9275.
Full textAbstract:
碩士國立臺北科技大學
電機工程系研究所
101
This paper presents an efficient real-time image stitching and localization algorithm. In the beginning, the blurred image frame is filtered out by considering the variation of its edge amount under different scales. Stitching a lot of frames with similar scene is time-consuming and easily generates the inconsistency in the overlap region between frames. In order to decrease the stitching error and the computational time, the proposed image stitching strategy desires to stitch only several significant frames instead of aligning all frames. The dominant frame with most of the visual content dissimilar to that in the previously stitched result can contribute a lot of new information and is worthy to be stitched. For rising the image quality, a seam planning algorithm is designed to eliminate the unclear ghost effect in the overlap region between frames is proposed. Not only the color difference but also the richness of the involved visual content are considered to find the optimal seam. The visual content is tried to be maximally reserved during eliminating the inconsistency in the overlap region. The inconsistency is reduced by cutting off the overlap region with blending colors along the selected seam. On the other hand, the correspondence between the consecutive frames can be rapidly matched by feature point tracking with uniformly drawing feature points. The captured image at every frame can then be real-time localized on the mosaicking image by evaluating the homography matrix of the consecutive frames with the estimated correspondence. Through the verification of several experiments, the proposed algorithms in this thesis have presented the efficiency in image stitching and localization with the real-time performance.
21
Ting-YuLin and 林庭宇. "Efficient Point Selection and Matching for Real-time ICP-based Visual Odometry." Thesis, 2018. http://ndltd.ncl.edu.tw/handle/j86pen.
Full textAbstract:
碩士國立成功大學
電機工程學系
105
Visual odometry (VO) or Simultaneous localization and mapping (SLAM) for camera poses estimation plays an important role in virtual object registration of augmented reality (AR) applications. One of the main challenges of the algorithm is to keep the tracking errors as small as possible for reducing the discrepancy between virtual and real objects which can be easily recognized in practical application. Moreover, it is hard to keep up with the camera frame rate while trying to maintain small error. Therefore, implementing a fast and robust visual odometry algorithm is pursued. The proposed VO algorithm is based on iterative closest point (ICP) algorithm, which is a widely used registration algorithm but the quality of ICP estimation is easily affected by insufficient structural constraint and noise. Several methods have been proposed to sample points with enough transformation constraints; however heavy computation is necessary for knowing normal vectors of every pixels. In this thesis, we propose to sample points from near-edge regions, which can effectively increase constraints by increasing the possibility of sampling points from various objects in the scene. Besides, to decrease the influence of axial noise, we only sample points from near-edge regions inside the region of interest (ROI) which is extended from the near-end of the view, and then is adaptively extended or shrunk to ensure sufficient constraints while minimizing the influence of axial noise. Secondly, we aim at reducing the complexity of matching stage in ICP. Based on multi-resolution scheme, an idea of adaptive searching area determination is proposed to reduce the redundant iterations as a smaller searching area is sufficient to search for true match in small camera motion. In addition, through using temporal correlation of ICP in steady state at the initial estimate for finding the closest point, number of search for each source points can be furthered reduced. Finally, the proposed scheme is extended to the keyframe-based method where a keyframe is generated when the distance between the current frame and the last keyframe is large. Howerver, since the distance is unknown before estimation, a poor matching quality may occur, which results in large error; Hence, a strategy of efficiently changing keyframe and then performing re-estimation is proposed to increase the matching quality. The proposed VO algorithm is evaluated on publicly available benchmark dataset. Compared with other VO algorithms, the proposed one exhibits competitive performance and achieves an average frame rate of 136 FPS using only a single CPU thread.
22
Hsu, Jung-Hsien, and 許榮顯. "Real-time Visual tracking Based Ball-and-Beam Balance Control System Design." Thesis, 2016. http://ndltd.ncl.edu.tw/handle/35y59g.
Full textAbstract:
碩士國立臺灣科技大學
電機工程系
104
This research designs a real-time visual tracking ball-and-beam system via FPGA. The structure of ball-and-beam balance control system includes image processing module and controller module. The system is implemented on FPGA and programming with Verilog. The system applies image sensor to capture the real-time ball-and-beam information, and calculates ball position and beam angle. In this research, we applied system identification, state space equation, PID control, and fuzzy logic control to design controllers. Control performance and FPGA implemented resources are compared from the experiment results. This thesis achieves both the single-point and multi-point ball position control on real-time visual tracking ball-and- beam system.
23
Chuang, Chia-Zuo, and 莊家碩. "A Real-Time Feature-Based Visual Tracking System for Hand-Eye Coordination." Thesis, 2000. http://ndltd.ncl.edu.tw/handle/29094200250722559922.
Full textAbstract:
碩士國立東華大學
電機工程研究所
88
Theoretical and practical issues in vision-based control systems is becoming a major research area in robotics to which considerable research effort has been devoted. Among those dedicated researchers, the unified framework for vision-based control systems proposed by Chang[[6]] has played an important role in analyzing such systems. The purpose of the research is to develop a real-time visual tracking technique based on the unified framework that can be applied to robotic hand-eye coordination. Moreover, precise robot positioning systems employing approximately calibrated stereo vision can thus be implemented. The two-camera vision systems together with the proposed visual tracking technique perform just like human eyes that provide a global measure for a robot in a changing environment. Although real-time visual tracking systems have been applied to fields like robotics and visual surveillance, it is still lack of a systematic approach for developing such systems. In the research, we propose a visual tracking systems employing feature tracking method under the aforementioned unified framework. This formulation enables the design of visual tracking systems to be performed in a fairly general setting and allows the objects to move freely. The feature-based tracking system proposed for hand-eye coordination is capable of tracking multiple lists of features in a systematic way. This means that this visual tracking system includes a framework for composing basic features to track an object, and therefore is capable of simultaneously tracking multiple features. Our complete system includes a two-camera vision system with pan-tilt heads and image digitizers, and a PC performing all computation for real-time visual tracking of multiple objects.
24
Silva, Gilberto Miguel Ribeiro. "Time-frequency and coherence based studies of the neural correlates of visual perception." Master's thesis, 2014. http://hdl.handle.net/10316/27993.
Full textAbstract:
Dissertação de Mestrado Integrado em Engenharia Biomédica apresentada à Faculdade de Ciências e Tecnologia da Universidade de CoimbraAo longo dos últimos anos, o estudo na área das neurociências tem vindo a focarse na compreensão da complexidade do cérebro humano; utilizando as mais variadas técnicas que, evoluindo ao longo do tempo, se transformam em potenciais “janelas” na exploração do mais complexo e menos conhecido órgão humano. Explorando a forte vantagem do electroencefalograma e a sua poderosa resolução temporal, o objectivo desta tese é aprofundar conhecimentos e construir métodos de análise das oscilações cerebrais, nomeadamente, o estudo das suas características em tempo e frequência. Actualmente são realizadas inúmeras referências à relação entre as oscilações e percepção, atenção, memória, aprendizagem, integração de informação, entre outras. A compreensão destes mecanismos pode ser inferida através das oscilações, pelo que existe uma crescente vontade em fazer uso da interpretação das mesmas para conhecimento dos diversos mecanismos cognitivos inerentes. No entanto, a classificação das oscilações em humanos é um tema complexo, e que carece de consenso no seio da comunidade científica. Por um lado, surge a dificuldade de relacionar os mecanismos na base da formação de tais oscilações, não sendo claras as fontes biológicas que contribuem para a sua formação. Por outro, temos as nuances na informação que, aparentemente, se relacionam com esta questão (sejam características genéticas ou até variabilidade inter-individual). No entanto, um princípio parece certo: a estabilidade das oscilações a nível individual. Nesta tese, valorizando uma análise automatizada e independente, serão abordados e implementados métodos de pré-processamento e análise “data-driven” para identificação de bandas em qualquer tipo de imagem “time-frequency”. Aplicando os algoritmos criados, será procurada relação entre bandas de oscilações entre grupos (controlo e indivíduos com neurofibromatose tipo-1) e entre sinais hemodinâmicos de BOLD e níveis de GABA
25
"EEG-Based Estimation of Human Reaction Time Corresponding to Change of Visual Event." Master's thesis, 2019. http://hdl.handle.net/2286/R.I.55526.
Full textAbstract:
abstract: The human brain controls a person's actions and reactions. In this study, the main objective is to quantify reaction time towards a change of visual event and figuring out the inherent relationship between response time and corresponding brain activities. Furthermore, which parts of the human brain are responsible for the reaction time is also of interest. As electroencephalogram (EEG) signals are proportional to the change of brain functionalities with time, EEG signals from different locations of the brain are used as indicators of brain activities. As the different channels are from different parts of our brain, identifying most relevant channels can provide the idea of responsible brain locations. In this study, response time is estimated using EEG signal features from time, frequency and time-frequency domain. Regression-based estimation using the full data-set results in RMSE (Root Mean Square Error) of 99.5 milliseconds and a correlation value of 0.57. However, the addition of non-EEG features with the existing features gives RMSE of 101.7 ms and a correlation value of 0.58. Using the same analysis with a custom data-set provides RMSE of 135.7 milliseconds and a correlation value of 0.69. Classification-based estimation provides 79% & 72% of accuracy for binary and 3-class classication respectively. Classification of extremes (high-low) results in 95% of accuracy. Combining recursive feature elimination, tree-based feature importance, and mutual feature information method, important channels, and features are isolated based on the best result. As human response time is not solely dependent on brain activities, it requires additional information about the subject to improve the reaction time estimation.Dissertation/Thesis
Masters Thesis Electrical Engineering 2019
26
Huang, Yen-Shiung, and 黃彥翔. "Real-Time Visual Tracking Techniques Based on Edge Detection Applied for Augmented Reality." Thesis, 2009. http://ndltd.ncl.edu.tw/handle/55874505533870954653.
Full textAbstract:
碩士國立臺灣科技大學
資訊工程系
97
In this thesis, we present real-time visual tracking techniques applied for augmented reality. The visual sensor employed is a common USB camera for capturing source images, and the objects tracked are the artificial square markers with particular patterns and given sizes. Our proposed method enables computers to sense each marker whether exists in the sight of the camera, which identity the marker has, and what pose and position it is. With this information, computers can superimpose virtual objects on the specific markers respectively, so that users can interact with the virtual objects simultaneously by interacting with the markers. The tracking method described in this thesis includes the quadrangle detection, marker recognition, and three-dimensional pose and position estimation. Our quadrangle detection method is based on the edge detection, and we use the edges with their skeletons and directions to search the vertices of quadrangles in the source image. In comparison with a threshold based method, our method is able to work in a wider range of the environmental lightness. The vertices are used to normalize the corresponding quadrangle for marker recognition and to estimate the spatial relations between the marker and the camera in the real world. The normalization process in our marker recognition method is not only to normalize the shape of the quadrangle but also to normalize the color of it for minimizing the errors of the marker identification. Each of our designed markers contains a 16-bit code which gives a good variety of identities to them. Our pose and position estimation method is camera-calibration free as long as the vertical angle of view and the focal length of the camera are given. According to the experimental results, the average processing time of a 320 × 240 image for tracking 5 markers is 9.4 milliseconds corresponding to 104.6 FPS, and it takes 32.7 milliseconds corresponding to 30.6 FPS on an average to process a 640 × 480 image for tracking the same number of markers with a 2.4 GHz, single-core processor. It means that our tracking techniques can track multiple objects in real time.
27
Chen, Shih-Ying, and 陳思穎. "Study of Human Visual Adaptation Based on Shader Technology in Real-Time 3D Rendering." Thesis, 2006. http://ndltd.ncl.edu.tw/handle/73351392900275735879.
Full textAbstract:
碩士國立臺北教育大學
玩具與遊戲設計研究所
94
Due to the fast development of hardware and graphic technologies, the transforming, lighting, and texturing in fix rendering pipeline can be replaced with programming shader used in some rendering graphic technologies and plentiful graphic effects in real-time 3D environment also can be achieved. In this study, a method which simulating the human visual system for lighting adaptation with the pupil of the eyes has been proposed. This study integrates the related graphic techniques, such as High Dynamic Range Image (HDRI)、tone reproduction and the human visual system, with the vertex shader and the pixel shader to simulate the lighting adaptation effect in the human visual system. A visual 3D rendering environment, built with vertex and pixel shader, which can produce effect with scenes changing in dark and light cases. This study simulates the dazzling light and the human visual system for light adapting with multiple-pass rendering technique. Comparing with the method proposed by Reinhard et al., this study improves the deviation in varieties of color, and separates the lighting adaptation range from the original system to achieve more real effect.
28
Huang, Hui-Pin, and 黃暉斌. "Design and Implementation of a Real-time Visual Tracking System Based on Multiple Cameras." Thesis, 2008. http://ndltd.ncl.edu.tw/handle/37043114660482676236.
Full textAbstract:
碩士國立成功大學
電機工程學系碩博士班
96
Intelligent visual surveillance has attracted increasing attention in recent years. The aim of this research is to use computer vision so that the system can automatically detect moving objects or track a specific target through a sequence of video frames. The system can also classify the target and analyze its action. Because the field of view (FOV) is limited when using a single camera, this paper adopts multiple cameras to implement a real-time tracking system. A modified adaptive background subtraction method is used to detect moving objects, and a multi-cue template matching approach is employed to track a moving target. The epipolar constraint and color histogram matching are exploited to deal with the correspondence problem that occurs when multiple cameras are used. In this thesis, an efficient and robust cooperation strategy is proposed to improve the tracking performance. The visual tracking system developed in this thesis can continuously track a moving target and estimate its position in other cameras even if the target is occluded in the FOV. The experimental results show that the proposed approach has satisfactory performance.
29
"Representing Sight and Sound in Design Media: A Cyclical Time-based Model." Doctoral diss., 2011. http://hdl.handle.net/2286/R.I.9166.
Full textAbstract:
abstract: As digital technology promises immediacy and interactivity in communication, sight and sound in motion graphics has expanded the range of design possibilities in advertising, social networking, and telecommunication beyond the visual realm. The experience of seeing has been greatly enriched by sound as visual solutions become dynamic and multi-dimensional. The ability to record and transfer sight and sound with new media has granted the designer more control in manipulating a viewer's experience of time and space. This control allows time-based form to become the foundation that establishes many interactive, multisensory and interdisciplinary applications. Is conventional design theory for print media adequate to effectively approach time-based form? If not, what is the core element that is required to balance the static and dynamic aspects of time in new media? Should time-related theories and methodologies from other disciplines be adopted into our design principles? If so, how would this knowledge be integrated? How can this experience in time be effectively transferred to paper? Unless the role of the time dimension in sight is operationally deconstructed and retained with sound, it is very challenging to control the design in this fugitive form. Time activation refers to how time and the perception of time can be manipulated for design and communication purposes. Sound, as a shortcut to the active time design element, not only encapsulates the structure of its "invisible" time-based form, but also makes changes in time conspicuously measurable and comparable. Two experiments reflect the influence of sound on imagery, a slideshow and video, as well as how the dynamics in time are represented across all design media. A cyclical time-based model is established to reconnect the conventional design principles learned in print media with time-based media. This knowledge helps expand static images to motion and encapsulate motion in stasis. The findings provide creative methods for approaching visualization, interactivity, and design education.Dissertation/Thesis
Experiment One
Experiment Two
Ph.D. Design 2011
30
林祐霆. "An One-Time Password Mutual Authentication Scheme Using Biometrics-Based Key and Visual Secret Sharing." Thesis, 2014. http://ndltd.ncl.edu.tw/handle/81436125917011100279.
Full textAbstract:
碩士逢甲大學
資訊工程學系
102
Nowadays, internet services are very popular so that users can get services by using not only their devices but also the devices which are not trusted or belong to their. However, the users’ information, which is used to authenticate may be stolen, when they use untrusted devices. Therefore, one-time password is more suitable in this environment. Considering the cost of authentication, the convenience for the users and link the authenticated information to the user directly, the proposed one-time password scheme nowadays may not be applicable for these requirements. In this research, we combine the peripheral what the users have, Visual Secret Sharing, and Biometrics-Based Key, to design a new one-time password scheme. This scheme not only can authenticate safely, but also have lower cost and directly link for the users. Then this scheme can confirm whether the login user is the legal real one or not.
31
Song, Xuan-qing, and 宋炫慶. "Real-Time Visual Detection and Tracking of Multiple Moving Objects Based on Particle Filter Techniques." Thesis, 2005. http://ndltd.ncl.edu.tw/handle/86592932642455768347.
Full textAbstract:
碩士國立臺灣科技大學
資訊工程系
93
In the last decade, due to the popularization of video products and the rapid development of computer vision techniques, the detection and tracking methods for dynamic images have been widely applied in many kinds of fields, such as video surveillance, intelligent transportation, and parking area management systems. They can replace a lot of bored and time-wasting work, and avoid mannal mistakes caused by fatigue of human. On the effectiveness for a given period of time, these visual detection and tracking systems possess the ability of reporting sudden situations in real time, so that the whole time costs of such systems can be greatly reduced. In this thesis, the detection phase of our developed system consists of four parts: background generation, foreground detection, shadow elimination, and background maintenance. In the background generation part, the median method is used for constructing background images from the past N frames. In the foreground detection part, an extraction function is applied to indirectly perform differencing to obtain foreground images. In the shadow elimination part, a deterministic nonmodel-based method is adopted to remove shadows. As to the background maintenance part, a history map which records the number of times of the changes of corresponding pixels is employed to maintain background images. In the tracking phase of the system, this thesis exploits a particle filter to track moving objects. The color distribution of a moving object is chosen as its features represented by a color probability histogram. In order to raise the accuracy of tracking, the background information serves as the increase candidate weight of a moving object. The experimental results reveal that in general situations our system can achieve real-time processing and can obtain robust detection and tracking results for multiple moving objects.
32
Yang, Tzong-Jer, and 楊宗哲. "On Real-Time 3D Head Motion Tracking and Its Applications in Model-Based Visual Communication." Thesis, 1999. http://ndltd.ncl.edu.tw/handle/02793968252520708780.
Full textAbstract:
博士國立臺灣大學
資訊工程學研究所
87
This dissertation addresses the problem of real-time 3D head motion tracking with a single camera. Through the analysis on a sequence of continuous facial images captured from a single video camera, the subject''s head motion in 3D space can be estimated. Two issues are discussed: feature tracking and 3D motion estimation. The purpose of feature tracking is to provide enough feature correspondence between two consecutive images, and the purpose of 3D motion estimation is to infer 3D motion from these feature correspondence. To tackle the motion-tracking problem, techniques of real-time feature tracking and 3D motion estimation are proposed. The proposed feature tracking technique is developed based on cross-correlation between two feature patterns, with additional constraints to obtain better tracking results. The proposed 3D motion estimation method includes knowledge of head motion to develop an objective function that is minimized iteratively to obtain motion parameters. Three feature points, two eye cor-ners and one nostril, are required, and the recovered motion parameters are three rota-tion angles and three translation offsets, respectively. The proposed techniques can be applied in different applications. One potential application is the model-based visual communication, where remote face images are synthesized based on parameters including those for facial expression and head mo-tion. Model-based coding is an emerging technique that can achieve very low bit-rate image coding, and is in particular applicable for mobile or network applications. A similar prototype system is developed in this dissertation. In the prototype, 3D head motion is inferred using the proposed techniques, and the recovered motion parame-ters are applied to a face synthesis module which renders a 3D texture-mapped head model in real-time. The techniques developed can not only be used in model-based visual communi-cation, but also be applied to applications requiring natural human-computer interface (HCI). A user can interact with computers or other intelligent appliances via head motions. The performance of the proposed 3D head motion tracking techniques is meas-ured on a PC with a Pentium-II 400 MHz CPU. The 3D motion estimation can update over 250 Hz, and the overall performance of the model-based visual communication system is over 25 frames/sec, with feature tracking and head model rendering in-volved.
33
Chuang, Chu-Lin, and 莊曲霖. "Study on Visual Surveillance-based Real-Time Detection and Encryption of Abandoned and Removed Objects." Thesis, 2013. http://ndltd.ncl.edu.tw/handle/41059718996674624920.
Full textAbstract:
碩士國立澎湖科技大學
電資研究所
101
This thesis proposes a visual surveillance-based real-time detection and encryption of abandoned and removed objects. In the proposed “Visual surveillance-based real-time detection of abandoned and removed objects”, the features of objects are used to detect the static foreground objects, and then make the obtained static foreground objects as candidated abandoned or removed objects. Singular value decomposition (SVD) is applied on the region of interest (ROI) of the candidated abandoned or removed object in the current frame and the background image, respectively. Next, the maximum eigenvalues of the S matrices are compared to obtain the abandoned object or the removed object. “Encryption and forgery detection of the abandoned and removed objects” is proposed to obtain the encryption and forgery detection of the abandoned or removed objects in the key-frames. The DWT+DCT+SVD-based image watermarking is used to obtain forgery detection by using the obtained features under embedding and extracting schemes. The proposed method can effectively detect the abandoned objects and the removed objects, and further determine whether images have been forged. The experimental results demonstrate that the proposed method has good performance. Furthermore, the false-positive detection in the SVD-based image watermarking is removed. Therefore, the proposed method has high value in the application of the visual surveillance.
34
Chen, Geng-Yen, and 陳庚延. "A real-time eyes detection system based on visual identification and its implementation on embedded systems." Thesis, 2011. http://ndltd.ncl.edu.tw/handle/5ufyvc.
Full textAbstract:
碩士國立臺北科技大學
資訊工程系研究所
99
Recently, there are many studies focus on the eye detection based on the computer vision technology. However, the eye detection system in the practical usage should work in the complex environment robustly and correctly. Besides, the efficiency of the eye detection system is the critical issue in order to achieve the system in real time. The eye detection system must overcome the complex background, the uneven brightness, the overall environmental illumination, and the different camera angles problems. These conditions will seriously affect the detection rate and detection time. Therefore, we present a low cost real-time eye detection which conquers the non-uniform and complex environmental illumination. The proposed framework can be divided into the following five technologies: the image enhancement technology, the reduction of the impact of uneven illumination technology, the face detection technology, and the eye detection technology. We use the Retinex algorithm to eliminate the impact of light, and then use the Haar-like features with AdaBoost learning algorithm to achieve face and eye detection. The proposed method is also implemented on the embedded system in real time. The experimental results demonstrate that the detection accuracy is over 90% with 10 ~ 20fps frame rate.
35
Chen, Chun-Chi, and 陳俊吉. "Visual-based Real-time Target Detection and Tracking in a Heterogeneous Dual-core Embedded Platform DSP." Thesis, 2015. http://ndltd.ncl.edu.tw/handle/58048467860682279624.
Full textAbstract:
碩士淡江大學
航空太空工程學系碩士班
103
This thesis mainly investigate the real-time target recognition and tracking using a heterogeneous dual-core embedded platform DSP. Recently DSP has wider and wider applications in the field of computer vision and image processing. Based on previous researches from our research group, this thesis integrate and expand their work. In this research, the ARM is in charge of exchanging data with other devices, including video device. These data is send to the DSP for further image processing. The processed results are transmitted back to the ARM for further applications. In most embedded devices, OpenCV is employed for image processing. However, the OpenCV is not compatible with the DSP in our case. Therefore, all the codes for image processing are developed by ourselves. The processed results are compared with those from OpenCV to ensure the correctness. Moreover, in the previous result the DSP only processes a single image. This thesis has reached the goal that DSP can process continuous image input. Experiments are also designed to check the correctness of the processed results. This research is potentially applicable to the real-time target recognition and tracking on board a UAV.
36
Feng, Yen-Chang, and 馮彥彰. "Design and Implementation of a Robust Real-Time Visual Tracking Algorithm Based on Multi-scale Classification." Thesis, 2016. http://ndltd.ncl.edu.tw/handle/55376161516734342206.
Full textAbstract:
碩士淡江大學
電機工程學系碩士班
104
Visual tracking acts an important role in fields of computer vision as it can apply on robots to accomplish many tasks. In this thesis, a robust visual tracking algorithm is proposed based on the existing compressive tracking method. The proposed algorithm adopts multiple Bayes feature classifiers, each of them was trained under a different scale condition, to realize online multi-scale classification, which can greatly improve the robustness of tracking system. Furthermore, each feature classifier was initialized by randomly generating different types of Haar-like features. By doing so, the robustness of feature classification can be improved to obtain more accurate tracking results. In order to enhance real-time performance of visual tracking system, the processing time of each stage of the proposed algorithm was recorded. Then, the processing speed of the proposed algorithm is accelerated by adopting CPU-based parallel processing on the computationally expensive stages. Also, the formula of the Bayes feature classifier is simplified by numerical analysis to speedup the processing speed of multi-scale feature classification. After accelerating by parallel computation and formula simplification, the proposed visual tracking algorithm reaches about 45 images per second (45 fps) tracking performance in dealing with images of 642x352 pixels. Experimental results show that the proposed algorithm outperforms four state-of-the-art visual tracking methods in terms of success rate, tracking accuracy, and visual comparison.
37
Lu, Cheng-Lun, and 呂政倫. "Study of a small boat-based multi-leading mark visual guidance approaching maneuver in the night time." Thesis, 2015. http://ndltd.ncl.edu.tw/handle/61947001118710020245.
Full textAbstract:
碩士國立臺灣海洋大學
通訊與導航工程學系
103
This work is concerned with the design and experiment of signal processing and fuzzy logic-based small boat automatic harbour entering control system in low visibility or night time condition. Typically, human pilots are preferred over autopilots during harbour entering maneuver because of their better adaptive capability and sensitivity to visual cues needed in ship handling. However, human pilots’ability to obtain reliable visual information is significantly reduced in low visibility conditions. Hence, machine vision-based system is sought to help or replace human pilots during night time harbor entering maneuvers. Specifically, a CCD camera mounted on the bow of an FRP boat is used to acquire images of the LED leading marks that define the leading line during harbour entry. The center of gravity of the captured images is computed via Labview/vision builder AI to find the deviated heading angle and the distance from the boat to the leading marks needed for the fuzzy autopilot to steer the boat along the leading lines. The limited emission angle of the LED lights causes nonuniform LED color emission, which might reduce the identifiability of the LED leading marks, and results in certain distance estimation error; hence, poor track-keeping performance. An Optical Diffuser Plate is placed in front of the LED light marks to produce uniform LED color emission and the signal processing-based distance estimation becomes more reliable. Successful experiments carried out at the National Taiwan Ocean University small boat harbor indicate that the proposed signal processing and fuzzy logic-based autopilot system is capable of guiding the boat along Z shapes routes defined by LED light marks in the night time.
38
Hu, Keyun. "VisArchive: A Time and Relevance Based Visual Interface for Searching, Browsing, and Exploring Project Archives (with Timeline and Relevance Visualization)." Thesis, 2014. http://hdl.handle.net/1828/5235.
Full textAbstract:
Project file archives are becoming increasingly large. The number of files, information and data that need to be created, accessed and modified throughout a project can be overwhelming. It is critical for project participants or contributors to find relevant information in project archives quickly. In this thesis, I present VisArchive, an interactive visualization tool that provides users with better awareness of search results within project archives. VisArchive visualizes the relevance-ranked search results with a color-coded stacked bar chart and interactive timelines and provides supporting visual cues to help differentiate search results based on searched keywords. It aims to allow users to interactively search, browse, and explore information in project archives, including access history, effectively and efficiently. I will present two case studies to illustrate how VisArchive can be used to support searching, browsing, and exploring information in building construction and open source software projects. In addition, I discuss how VisArchive can be improved to address information retrieval problems and work across different domains. VisArchive demonstrates the combination and application of several visualization techniques to the problem of searching and navigating project archives.Graduate
0984
39
Ahmad, J., M. Sajjad, Irfan Mehmood, S. Rho, and S. W. Baik. "Saliency-weighted graphs for efficient visual content description and their applications in real-time image retrieval systems." 2015. http://hdl.handle.net/10454/17187.
Full textAbstract:
YesThe exponential growth in the volume of digital image databases is making it increasingly difficult to retrieve relevant information from them. Efficient retrieval systems require distinctive features extracted from visually rich contents, represented semantically in a human perception-oriented manner. This paper presents an efficient framework to model image contents as an undirected attributed relational graph, exploiting color, texture, layout, and saliency information. The proposed method encodes salient features into this rich representative model without requiring any segmentation or clustering procedures, reducing the computational complexity. In addition, an efficient graph-matching procedure implemented on specialized hardware makes it more suitable for real-time retrieval applications. The proposed framework has been tested on three publicly available datasets, and the results prove its superiority in terms of both effectiveness and efficiency in comparison with other state-of-the-art schemes.
Supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (2013R1A1A2012904).
40
"Image Processing Based Control of Mobile Robotics." Master's thesis, 2016. http://hdl.handle.net/2286/R.I.40366.
Full textAbstract:
abstract: Toward the ambitious long-term goal of a fleet of cooperating Flexible Autonomous Machines operating in an uncertain Environment (FAME), this thesis addresses various control objectives for ground vehicles.
There are two main objectives within this thesis, first is the use of visual information to control a Differential-Drive Thunder Tumbler (DDTT) mobile robot and second is the solution to a minimum time optimal control problem for the robot around a racetrack.
One method to do the first objective is by using the Position Based Visual Servoing (PBVS) approach in which a camera looks at a target and the position of the target with respect to the camera is estimated; once this is done the robot can drive towards a desired position (x_ref, z_ref). Another method is called Image Based Visual Servoing (IBVS), in which the pixel coordinates (u,v) of markers/dots placed on an object are driven towards the desired pixel coordinates (u_ref, v_ref) of the corresponding markers.
By doing this, the mobile robot gets closer to a desired pose (x_ref, z_ref, theta_ref).
For the second objective, a camera-based and noncamera-based (v,theta) cruise-control systems are used for the solution of the minimum time problem. To set up the minimum time problem, optimal control theory is used. Then a direct method is implemented by discretizing states and controls of the system. Finally, the solution is obtained by modeling the problem in AMPL and submitting to the nonlinear optimization solver KNITRO. Simulation and experimental results are presented.
The DDTT-vehicle used within this thesis has different components as summarized below:
(1) magnetic wheel-encoders/IMU for inner-loop speed-control and outer-loop directional control,
(2) Arduino Uno microcontroller-board for encoder-based inner-loop speed-control and encoder-IMU-based outer-loop cruise-directional-control,
(3) Arduino motor-shield for inner-loop speed-control,
(4) Raspberry Pi II computer-board for outer-loop vision-based cruise-position-directional-control,
(5) Raspberry Pi 5MP camera for outer-loop cruise-position-directional control.
Hardware demonstrations shown in this thesis are summarized: (1) PBVS without pan camera, (2) PBVS with pan camera, (3) IBVS with 1 marker/dot, (4) IBVS with 2 markers, (5) IBVS with 3 markers, (6) camera and (7) noncamera-based (v,theta) cruise control system for the minimum time problem.Dissertation/Thesis
Masters Thesis Electrical Engineering 2016
41
Hsu, Fang-Chin, and 徐芳秦. "Spatial-Aware Channel Attention Mechanism and Temporal Embedding Propagation Module Based on Deep Learning Architecture for Real-time Visual Object Tracking System." Thesis, 2019. http://ndltd.ncl.edu.tw/handle/t2vajv.
Full textAbstract:
碩士國立臺灣科技大學
電機工程系
107
This study proposed a spatial-aware channel attention mechanism and temporal embedding propagation module based on deep learning architecture for real-time visual object tracking system. The proposed method is an advanced research based on the SiamRPN, and it is an end-to-end off-line trained network, and can provide real-time efficiency. For the proposed spatial-aware channel attention mechanism, we apply it to reweight the channels while extracting the template feature and enhance the generalization power of the model. The system can benefit from the mechanism and find the representative feature which can adapt to the time-varied appearance of an object or background. For the proposed temporal embedding propagation module, we design it to address the scale changes and deformation problem. In our method, the module can efficiently utilize the advantage of temporal information between the adjacent frames. The object information in the previous frame can be transformed into a single dimension embedding vector using our designed RoI embedding layer and propagate to the current frame. The operation can provide the current feature of the target object, and can also increase the discriminative power in the similarity stage. This study conducts several of experiments on the public benchmark OTB and the VOT challenges, and is compared against previous works. Though there are uncontrolled factors in the videos, such as illumination changes and fast motion, the proposed method can achieve superior accuracy than the former schemes. Thus, the proposed method has considerable potential to be applied in the practical applications.
42
Wang, Chun-Kai, and 王俊凱. "Design and Implementation of a Multi-Purpose Real-time Visual Tracking System based on Modified Adaptive Background Subtraction and Multi-Cue Template Matching." Thesis, 2004. http://ndltd.ncl.edu.tw/handle/64376820697850240847.
Full textAbstract:
碩士國立成功大學
電機工程學系碩博士班
92
How to detect and track moving objects in video streams is an important and challenging research problem in visual tracking applications. Among the moving object detection algorithms, the background subtraction method has been widely used due to its less computational load and high detection quality. Nevertheless, with problems such as “sensitive to the changes of lighting” and “background initialization”, the background subtraction method often performs poorly in complex environments. To overcome this difficulty, a modified adaptive background subtraction method that can dynamically update the background model is proposed. On the other hand, the SSD method is a popular image tracking technique in visual servoing applications. However, the SSD method is very sensitive to the changes of target appearance. As a consequence, it may experience failures when applied to realistic environments. To overcome this difficulty, a multi-cue template matching algorithm that consists of several kinds of tracking algorithms is proposed to improve the practicability and robustness of the visual tracking system. To evaluate the performance of the proposed approach, the modified adaptive background subtraction method and the multi-cue template matching algorithm are applied to the real-time visual tracking system developed in our laboratory. Experimental results show that the proposed approach exhibits satisfactory performances.