Rapid Detection And Treatment Analysis Of Tunnel Lining Damage Based On Deep Learning

As the length of roads and railways increases,the roads will have to be extended to the natural environment away from the cities.Since the reform and opening up,the number and scale of China's tunnels are continuously increasing,but the problems of various diseases in the operation period also come,and has become one of the important problems facing the tunnels.In order to ensure the safety of tunnel operation and improve the service life of tunnels,many detection methods have been proposed in the field of tunnel detection,such as traditional manual detection method,ultrasonic detection method,ground-penetrating radar method,laser scanning method and inspection method based on image processing technology.However,due to the high cost of equipment,single test content,strict test environment and other reasons,most of the current tunnel routine inspection is still manual inspection.This paper proposes a deep-learning-based rapid detection and treatment analysis method for tunnel lining damage.The main results of the applied research are as follows:(1)Through literature review and field investigation,this paper selects three common diseases of tunnel lining: lining crack,lining leakage,lining freezing damage,and divides the three diseases into seven types according to their manifestations.Then these diseases are analyzed in detail,including their common location,causes and treatment measures.(2)Based on deep learning,the AlexNet convolutional neural network was used to train,verify and test 300,000 disease images.The test results show that the convolutional neural network can realize the fast and accurate identification of lining diseases.(3)Based on MATLAB GUI,the damage detection program of tunnel lining is designed.The program uses man-machine interaction to extract the length,width and leakage area of cracks.By combining the research results of convolutional neural network and disease analysis with the program,a one-stop solution for the whole process of tunnel lining detection,identification,damage extraction,disease analysis and treatment is designed.The test results show that the detection method and program presented in this paper perform well and will provide better help for tunnel detection after many debugging and optimization.