Project Title	AI-based Data Analytics for Pavement Distress Classification using Video Data
University	Virginia Tech University
Principal Investigator(s)	Linbing Wang
Funding Source(s) and Amounts Provided (by each agency or organization)	Federal Funds, $49,672 Match $49,712
Total Project Cost	$99,384
Start and End Dates	09/01/2022 - 09/31/2023
Brief Description of Research Project	The long-term goal of the research is to develop a high accuracy, high speed pavement distress type classification system and cellphone application using video data and smart phone camera live streaming capabilities. Using video input, the pavement distress detection system will be able to identify specific pavement distresses including potholes, alligator cracking, longitudinal cracking, transverse cracking, edge cracking, patching, and rutting. Particularly, the study has the following sub-objectives: 1) To provide a comprehensive review of distress classification and detection based on AI approaches. 2) To improve the machine learning methods on pavement distress detection. 3) To establish a pavement distress image dataset for machine learning training and testing of over 10,000 images. 4) To develop a comprehensive system that can automatically detect pavement distress in images, videos, and real-time video live streams. 5) To develop a cellphone application that can integrate the detection system and make distress detection easier for both expert and novice.