Computer Vision-based Studies for Autonomous Condition Assessment of Civil Infrastructure Systems
|Interdisciplinary Areas:||Data/Information/Computation, Smart City, Infrastructure, Transportation
This project will lead to advancements in machine learning and computer vision approaches for detection, quantification and localization of defective regions on civil infrastructure systems. Regular inspections and maintenance of infrastructure systems will prolong their service life. Visual manual assessment is currently the predominant method used for the inspection of most infrastructures which is a subjective, time-consuming, and labor-intensive process that is highly prone to human error. There is an urgent need to develop more effective approaches for the inspection of infrastructure systems – these include using machine learning, computer vision and robotic systems. The postdoctoral researcher will develop new approaches for autonomous analysis of image-based data for condition assessment of civil infrastructure systems (e.g., buildings, roads, dams, bridges, nuclear power plants, sewer pipelines, etc.). The selected candidate will utilize state-of-the-art facilities at Purdue’s Video and Image Processing Laboratory (VIPER) and Smart Informatix Laboratory as well as Bowen Laboratory to develop filed application solutions to uplift the condition of infrastructures by facilitating the next generation of quantitative and more frequent inspections and health monitoring.
PhD in Electrical or Computer Engineering, Civil Engineering, or closely related fields with strong background and experience in Computer Vision, Deep Learning, and Machine Learning.
Mohammad R. Jahanshahi
Assistant Professor of Civil Engineering
Smart Informatix Laboratory
The Charles William Harrison Distinguished Professor of Electrical and Computer Engineering
Professor of Biomedical Engineering
Professor of Psychological Sciences (Courtesy)
Video and Image Processing Laboratory (VIPER)
1. Fu-Chen Chen and Mohammad R. Jahanshahi, (2018), "NB-CNN: Deep learning-based crack detection using convolutional neural network and naive Bayes data fusion," IEEE Transactions on Industrial Electronics, Vol. 65, No. 5, May 2018, 4392-4400, DOI: 10.1109/TIE.2017.2764844.