Wu, Weidong; Owino, Joseph
Fomunung, Ignatius; Onyango, Mbakisya A.; Liang, Yu
College of Engineering and Computer Science
University of Tennessee at Chattanooga
Place of Publication
In this thesis, a real-time and low-cost solution to the autonomous condition assessment of pavement is proposed using deep learning, Unmanned Aerial Vehicle (UAV) and Raspberry Pi tiny computer technologies, which makes roads maintenance and renovation management more efficient and cost effective. A comparison study was conducted to compare the performance of seven different combinations of meta-architectures for pavement distress classification. It was observed that real-time object detection architecture SSD with MobileNet feature extractor is the best combination for real-time defect detection to be used by tiny computers. A low-cost Raspberry Pi smart defect detector camera was configured using the trained SSD MobileNet v1, which can be deployed with UAV for real-time and remote pavement condition assessment. The preliminary results show that the smart pavement detector camera achieves an accuracy of 60% at 1.2 frames per second in raspberry pi and 96% at 13.8 frames per second in CPU-based computer.
First and foremost, I would like to express my thanks to the Almighty Allah for providing me a golden opportunity and healthy life to pursue a master’s degree on Artificial Intelligence (AI) using UAV at the University of Tennessee at Chattanooga (UTC), my deepest appreciation and humbleness to Him. Thanks to my parents, my beautiful wife, my lovely son, and my family members to support me to pursue master’s degree without any delay. They support me in mentally as well as financially to go through the educational system of the USA. I am also grateful to Allah to have Dr. Weidong Wu as my academic advisor and coordinator. Dr. Wu is a nice and easy access person with good knowledge of advanced technologies in civil engineering, and I am thankful to him for his supervision, support, guidance, constructive criticism, and encouragement. Also, my gratitude is will go to Dr. Owino, Dr. Fomunung, Dr. Onoyango, and Dr. Liang for their intellectual support and critical criticism of my thesis manuscript. Moreover, my gratitude to Dr. Arash for his advice and good recommendation to pursue my PhD study. I would like to thank TN Board of Architectural and Engineering Examiners (TBAEE) and University of Tennessee at Chattanooga (UTC) for funding support. Supercomputer resources used for deep learning simulations provided by Drs Anthony Skjellum and Ethan Hereth and UTC SimCenter are greatly appreciated. I would like to thankful to Mrs. Lemon, Mrs. Anderson, Mrs. Cook, and other staffs of the Department of Civil and Chemical Engineering for their support and encouragement of my study at UTC. I thank my fellow lab-mate Mr. Babatunde Atolagbe for the encouragement, for the sleepless nights we worked together, and all the fun that we made during 2 years of long journey. Moreover, my special thanks to Mawazo Fortonatos, Abubakr Ziedan, Sayed Mohammad Tarek, Shuvasish Roy, Maxwell Omwenga, and Raiful Hasan for their support in on-campus or off-campus. Last but not the least, I would thankful to Dr. Richard Wilson; Clarence Francis to make my stay in the U.S. memorable and enjoyable. Their support cannot measure in terms of money.
M. S.; A thesis submitted to the faculty of the University of Tennessee at Chattanooga in partial fulfillment of the requirements of the degree of Master of Science.
Pavements -- Design and construction
xv, 89 leaves
Qurishee, Murad Al, "Low-cost deep learning UAV and Raspberry Pi solution to real time pavement condition assessment" (2019). Masters Theses and Doctoral Dissertations.