Kisacanin / Singh / Bhattacharyya Embedded Computer Vision

1. Auflage 2008
ISBN: 978-1-84800-304-0
Verlag: Springer
Format: PDF
Kopierschutz: 1 - PDF Watermark

E-Book, Englisch, 284 Seiten

Reihe: Advances in Computer Vision and Pattern Recognition

As a graduate student at Ohio State in the mid-1970s, I inherited a unique c- puter vision laboratory from the doctoral research of previous students. They had designed and built an early frame-grabber to deliver digitized color video from a (very large) electronic video camera on a tripod to a mini-computer (sic) with a (huge!) disk drive-about the size of four washing machines. They had also - signed a binary image array processor and programming language, complete with a user's guide, to facilitate designing software for this one-of-a-kindprocessor. The overall system enabled programmable real-time image processing at video rate for many operations. I had the whole lab to myself. I designed software that detected an object in the eldofview,trackeditsmovementsinrealtime,anddisplayedarunningdescription of the events in English. For example: 'An object has appeared in the upper right corner...Itismovingdownandtotheleft...Nowtheobjectisgettingcloser...The object moved out of sight to the left'-about like that. The algorithms were simple, relying on a suf cient image intensity difference to separate the object from the background (a plain wall). From computer vision papers I had read, I knew that vision in general imaging conditions is much more sophisticated. But it worked, it was great fun, and I was hooked.

Autoren/Hrsg.

Singh, Sameer

Kisa?anin, Branislav

Chai, Sek

Kisacanin, Branislav

Bhattacharyya, Shuvra S.

Inhaltsverzeichnis

1;Foreword;6
2;Preface;8
2.1;Embedded Computer Vision;8
2.2;Target Audience;9
2.3;Organization of the Book;10
2.4;Overview of Chapters;10
2.5;How This Book Came About;12
2.6;Outlook;13
3;Acknowledgements;14
4;Contents;15
5;List of Contributors;22
6;Introduction;26
6.1;Hardware Considerations for Embedded Vision Systems;27
6.1.1;1.1 The Real-Time Computer Vision Pipeline;27
6.1.2;1.2 Sensors;29
6.1.3;1.3 Interconnects to Sensors;33
6.1.4;1.4 Image Operations;35
6.1.5;1.5 Hardware Components;36
6.1.6;1.6 Processing Board Organization;46
6.1.7;1.7 Conclusions;48
6.1.8;References;49
6.2;Design Methodology for Embedded Computer Vision Systems;51
6.2.1;2.1 Introduction;51
6.2.2;2.2 Algorithms;54
6.2.3;2.3 Architectures;55
6.2.4;2.4 Interfaces;57
6.2.5;2.5 Design Methodology;59
6.2.6;2.6 Conclusions;67
6.2.7;References;67
6.3;We Can Watch It for You Wholesale;72
6.3.1;3.1 Introduction to Embedded Video Analytics;72
6.3.2;3.2 Video Analytics Goes Down-Market;74
6.3.3;3.3 How Does Video AnalyticsWork?;79
6.3.4;3.4 An Embedded Video Analytics System: by the Numbers;89
6.3.5;3.5 Future Directions for Embedded Video Analytics;93
6.3.6;3.6 Conclusion;97
6.3.7;References;98
7;Advances in Embedded Computer Vision;100
7.1;Using Robust Local Features on DSP-Based Embedded Systems;101
7.1.1;4.1 Introduction;101
7.1.2;4.2 RelatedWork;103
7.1.3;4.3 Algorithm Selection;104
7.1.4;4.4 Experiments;109
7.1.5;4.5 Conclusion;119
7.1.6;References;121
7.2;Benchmarks of Low-Level Vision Algorithms for DSP, FPGA, and Mobile PC Processors;123
7.2.1;5.1 Introduction;123
7.2.2;5.2 RelatedWork;125
7.2.3;5.3 Benchmark Metrics;125
7.2.4;5.4 Implementation;126
7.2.5;5.5 Results;139
7.2.6;5.6 Conclusions;140
7.2.7;References;141
7.3;SAD-Based Stereo Matching Using FPGAs;143
7.3.1;6.1 Introduction;143
7.3.2;6.2 RelatedWork;144
7.3.3;6.3 Stereo Vision Algorithm;145
7.3.4;6.4 Hardware Implementation;147
7.3.5;6.5 Experimental Evaluation;151
7.3.6;6.6 Conclusions;159
7.3.7;References;159
7.4;Motion History Histograms for Human Action Recognition;161
7.4.1;7.1 Introduction;161
7.4.2;7.2 RelatedWork;163
7.4.3;7.3 SVM-Based Human Action Recognition System;164
7.4.4;7.4 Motion Features;165
7.4.5;7.5 Dimension Reduction and Feature Combination;170
7.4.6;7.6 System Evaluation;172
7.4.7;7.7 FPGA Implementation on Videoware;178
7.4.8;7.8 Conclusions;182
7.4.9;References;183
7.5;Embedded Real-Time Surveillance Using Multimodal Mean Background Modeling;185
7.5.1;8.1 Introduction;185
7.5.2;8.2 RelatedWork;186
7.5.3;8.3 Multimodal Mean Background Technique;188
7.5.4;8.4 Experiment;190
7.5.5;8.5 Results and Evaluation;192
7.5.6;8.6 Conclusion;196
7.5.7;References;197
7.6;Implementation Considerations for Automotive Vision Systems on a Fixed- Point DSP;198
7.6.1;9.1 Introduction;198
7.6.2;9.2 Fixed-Point Arithmetic;203
7.6.3;9.3 Process of Dynamic Range Estimation;203
7.6.4;9.4 Implementation Considerations for Single-Camera Steering Assistance Systems on a Fixed- Point DSP;207
7.6.5;9.5 Results;211
7.6.6;9.6 Conclusions;214
7.6.7;References;215
7.7;Towards OpenVL: Improving Real-Time Performance of Computer Vision Applications;216
7.7.1;10.1 Introduction;216
7.7.2;10.2 RelatedWork;218
7.7.3;10.3 A Novel Software Architecture for OpenVL;222
7.7.4;10.4 Example Application Designs;232
7.7.5;10.5 Conclusion and Future Work;235
7.7.6;10.6 Acknowledgements;236
7.7.7;References;236
8;Looking Ahead;238
8.1;Mobile Challenges for Embedded Computer Vision;239
8.1.1;11.1 Introduction;239
8.1.2;11.2 In Search of the Killer Applications;241
8.1.3;11.3 Technology Constraints;244
8.1.4;11.4 Intangible Obstacles;250
8.1.5;11.5 Future Direction;252
8.1.6;References;253
8.2;Challenges in Video Analytics;256
8.2.1;12.1 Introduction;256
8.2.2;12.2 Current Technology and Applications;257
8.2.3;12.3 Building Blocks;263
8.2.4;12.4 Embedded Implementations;267
8.2.5;12.5 Future Applications and Challenges;269
8.2.6;12.6 Summary;273
8.2.7;References;274
8.3;Challenges of Embedded Computer Vision in Automotive Safety Systems;276
8.3.1;13.1 Computer Vision in Automotive Safety Applications;276
8.3.2;13.2 Literature Review;277
8.3.3;13.3 Vehicle Cueing;278
8.3.4;13.4 Feature Extraction;287
8.3.5;13.5 Feature Selection and Classification;293
8.3.6;13.6 Experiments;295
8.3.7;13.7 Conclusion;297
8.3.8;References;297
9;Index;299