Improved road crack detection model based on U-Net network
Author:Haolu Li Author Unit: Haolu Li，Shigang Hu
Abstract：With the development of deep learning. Developing a neural network model for road crack detection is of great significance for improving peoples lives. With the rapid development of deep learning technology, semantic segmentation is widely used in road crack detection task. This paper introduces an improved model based on U-Net network. Firstly, the used dataset is introduced, and then the data preprocessing method is summarized, and then the road crack segmentation model is constructed, and finally the constructed crack segmentation model is trained. And we analyze the data of the trained road crack detection model.
Keywords：road crack detection; Deep learning; U-Net network; Model training
School of Information and Electrical Engineering, Hunan University of Science and Technology, Xiangtan 411201, China PDF
