Table of contents
Certificate of Original Authorship i
Certificate of Dissertation - Advisor’s consent iii
Certificate of Dissertation - Program Chair v
Acknowledgments vii
Abstract ix
List of figures xi
List of tables xiv
1 Introduction1
1.1 Objective of the Study 1
1.2 Limitations 2
2 Literature Review3
2.1 Pavement Defects3
2.1.1 Cracking 3
2.2 Autonomous Surveillance Vehicle6
2.3 Stereo Vision8
2.3.1 Methodology 8
2.3.2 Epipolar Geometry 9
2.3.3 Stereo Image rectification 10
2.3.4 Correspondence problem 11
2.3.5 Depth Calculation 12
2.4 Computer Vision and Machine Learning 13
2.4.1 CNN in Deep Learning 14
2.4.2 What is ANN14
2.4.3 CNN18
3 Methodology20
4 Data Acquisition and Results 22
4.1 Equipment 22
4.2 Softwares Used 23
4.3 Equipment setup 24
4.3.1 Data collection 25
4.3.2 Calibration 25
4.3.3 Disparity map26
4.3.4 Object Recognition / Crack Detection27
4.4 Results 29
4.4.1 Drawbacks 30
4.5 Integration with GPS33
4.6 Study Area34
4.7 Drawbacks 37
4.8 Future Scope37
References xxxviii
Appendix 1: Python Codes xl
Plagiarism Report Copy 46