Auto-detection of potholes using photogrammetric and deep learning approach (Softcopy is also available)
Datir, Raunak (PG190733)
Auto-detection of potholes using photogrammetric and deep learning approach (Softcopy is also available) - 2021 - xiii,52p.
Contents
Certificate of original authorship i
Certificate of Dissertation Advisor’s consent iii
Acknowledgments v
Abstract vii
List of figures ix
List of tables xii
1 Introduction 1
1.1 Aim and Objective 1
1.2 Thesis motivation 1
1.3 Significance of the study 2
1.4 Limitation 2
1.5 Thesis outline 3
2 Literature Review 5
2.1 Road surface Distress 5
2.1.1 Crack 5
2.1.2 Raveling 5
2.1.3 Roughness 6
2.1.4 Rutting 6
2.1.5 Potholes 7
2.2 Road Distress Detection Techniques 7
2.2.1 Public Reporting 8
2.2.2 Vibration based techniques 8
2.2.3 2D and 3D vision based technique 8
2.2.4 Learning based technique 10
2.3 Inferences 12
3 Methodology 13
4 Data Collection and Processing 15
4.1 Camera Module 15
4.2 Testing 17
4.2.1 Testing without Ground Control Points (GCPs) 17
4.2.2 Testing with Ground Control Points (GCPs) 19
4.3 Study area and dataset 23
4.3.1 Open source dataset 25
5 Results and Conclusion 27
5.1 Photogrammetric approach 27
5.1.1 Steps for data processing 27
5.2 Deep learning approach 30
5.2.1 Mask RCNN 31
5.2.2 YOLO (You only look once) 35
5.3 Conclusion 38
5.4 Limitation and Drawbacks 39
5.5 Future Scope 39
References xli
Appendix 1: Various test outputs to improve accuracy of camera module xlv
Appendix 2: Python code xlvii
Plagiarism Report copy li
MG TH-0228 / DAT
Auto-detection of potholes using photogrammetric and deep learning approach (Softcopy is also available) - 2021 - xiii,52p.
Contents
Certificate of original authorship i
Certificate of Dissertation Advisor’s consent iii
Acknowledgments v
Abstract vii
List of figures ix
List of tables xii
1 Introduction 1
1.1 Aim and Objective 1
1.2 Thesis motivation 1
1.3 Significance of the study 2
1.4 Limitation 2
1.5 Thesis outline 3
2 Literature Review 5
2.1 Road surface Distress 5
2.1.1 Crack 5
2.1.2 Raveling 5
2.1.3 Roughness 6
2.1.4 Rutting 6
2.1.5 Potholes 7
2.2 Road Distress Detection Techniques 7
2.2.1 Public Reporting 8
2.2.2 Vibration based techniques 8
2.2.3 2D and 3D vision based technique 8
2.2.4 Learning based technique 10
2.3 Inferences 12
3 Methodology 13
4 Data Collection and Processing 15
4.1 Camera Module 15
4.2 Testing 17
4.2.1 Testing without Ground Control Points (GCPs) 17
4.2.2 Testing with Ground Control Points (GCPs) 19
4.3 Study area and dataset 23
4.3.1 Open source dataset 25
5 Results and Conclusion 27
5.1 Photogrammetric approach 27
5.1.1 Steps for data processing 27
5.2 Deep learning approach 30
5.2.1 Mask RCNN 31
5.2.2 YOLO (You only look once) 35
5.3 Conclusion 38
5.4 Limitation and Drawbacks 39
5.5 Future Scope 39
References xli
Appendix 1: Various test outputs to improve accuracy of camera module xlv
Appendix 2: Python code xlvii
Plagiarism Report copy li
MG TH-0228 / DAT