Published: 
By  Sensing and Evaluation Laboratory (I-S2EE)

Semantic Segmentation for Pixel-Level Crack Detection

Congratulations to Mohamad Alipour for getting our paper accepted for publication in the Journal for Computing in Civil Engineering. This paper, titled “Robust Pixel-Level Crack Detection Using Deep Fully Convolutional Neural Networks”, represents a novel approach to detecting localized cracking at the pixel level, a critical advancement to visual inspection of civil infrastructure. The work was developed in collaboration with Greg Miller at the University of Washington during Mohamad's summer research experience at UW.
Reference: Alipour, M., Harris, D.K., and Miller, G. (2019 accepted). “Robust Pixel-Level Crack Detection Using Deep Fully Convolutional Neural Networks.” Journal for Computing in Civil Engineering.