TOPIC
Cracking is one of the most serious distresses on transportation infrastructures (road, bridge, high-speed railways, etc.). The traditional method is mainly manual inspection by experienced transportation engineers, while in recent years, deep learning methods have been widely used as powerful tools to automatically detect and recognize the cracks on transportation infrastructures for timely repair and effective maintenance. In this seminar, I will briefly introduce our work on intelligent recognition of cracks on transportation infrastructures with limited dataset using three different approaches: a combined image processing algorithm + deep learning method; a lightweight designed Generative Adversarial Network-based method; and a lightweight designed few-shot learning method.
ABOUT THE LECTURER
Dr Yue Hou is a Senior Lecturer in Transportation Engineering of the Department of Civil Engineering of Faculty of Science and Engineering at Swansea University, UK. Yue’s research interests are focused on the resilient and intelligent transportation infrastructure and materials, physics-informed machine learning in transportation infrastructure engineering, design of sustainable infrastructure and materials using digital twins, and energy harvesting on transportation infrastructures. Yue is a Fellow of the Institution of Civil Engineers (FICE), a Fellow of the Institute of Physics (FInstP), a Fellow of the Institution of Engineering and Technology (FIET), IEEE Senior Member, and a Chartered Engineer (CEng). Yue’s research in intelligent transportation engineering as PI and Principal Researcher is widely supported by funding agencies in the UK, China, and US, including Natural Environmental Research Council (NERC), the National Natural Science Foundation of China (NSFC), Ministry of Science and Technology (MOST) of China, Federal Highway Administration (FHWA) of USA, etc., leading to production of more than 80 peer-reviewed papers, more than 40 invited presentations (talks) and more than 10 national patents applications.