Study On Tunnel Deformation Monitoring Using Digital Image Processing Techniques

Tunnel traffic, because of its adventage of shortening distances and less impact on the environment, it is widely used in urban subway, through the mountains or cross-river rail, road and water works. However, due to the complex geological conditions in the tunnel environment, tunnel construction disasters occur, causing heavy casualties and economic losses. Existing monitoring methods restrained by site conditions, often cannot reflect the tunnel deformation or easy application well. With the development of science and technology, the resolution of digital image recognition technology is improved continually and used more widely. In this paper, the theory and method of monitoring tunnel deformation by digital image processing technology is discussed, the main works are as follows.(1)The currently development and research status of monitoring tunnel deformation methods at home and abroad were described in detail, pointing out the limitations of current monitoring methods; at the same time, analysing the development, research status and the problems of digital image processing technology used in the tunnel deformation monitoring detailly, this paper presents main content of digital image processing technology to monitor tunnel deformation.(2)Discusses the theory of digital image processing techniques briefly,then discusses two image sub-pixel location technology, the gravity and fitting method, and wrote the target landmarks image sub-pixel positioning procedures, and through the experimental data to verify the reliability and accuracy of pixel coordinates extraction; describe the principles of digital image processing technology to monitor tunnel deformation method, the impact of lens distortion and pixel calibration method.(3)Uses the camera to explore a series of studies, compares the focus of sub-pixel positioning method and fitting method, draw the focus of high precision positioning method; the impact of lens distortion on the target resolution is not obvious; the length has little effect on the calibration target resolution, so it can be ignored; while certain shooting distance is known, the target resolution and the focal length are an inverse relationship. (4)By simulating the environment inside the tunnel and monitoring experiment, under certain conditions, the precision of using digital image processing technology to monitor tunnel deformation method is less than±0.30mm, and meets<code for urban rail transit engineering survey>(GB50308-2008), which horizontal displacement is±3mm and the vertical displacement is±0.5mm.Through in-depth theoretical and experimental studies, we can see the use of digital image processing technology to monitor tunnel deformation is feasible and high precision.The method also exists shortcomings:In the actual tunnel complex environment, it should ensure the stability of the imaging equipment installation, and keep the imaging plane parallel to the tunnel section.