Surrounding Rock Deformation Measurement Of Tunnel Based On Image Processing

In order to adapt to the rapid development of the national economy and the acceleration of urbanization process,under the historical opportunity of the National Highway Network Planning(2013-2030),the Medium and Long-term Railway Network Planning,the Western Development Strategy and the "One Belt and One Road" initiative,China has invested more and more in the public construction fields such as railways,highways and urban tracks.As one of the key links,the construction and operation of tunnels become the precondition to ensure smooth roads.However,as the tunnel construction sites are mostly complex geological areas,it is easy to cause deformation and collapse in construction and maintenance.Therefore,the measurement of tunnel deformation can effectively reduce the losses caused by disasters.Considering the shortcomings of the existing measurement methods,this paper improves the hardware design and software algorithm,studies a tunnel surrounding rock deformation measurement system based on image processing,which has automatic measurement,taking into account the measurement accuracy and high cost-effective ratio.On the hardware side,the laser emitter is fixed at the tunnel deformation monitoring point to form a clear light spot on the well target by designing red and blue lines to form the well target as a means to reduce camera distortion.The "well" target image with spot is collected through industrial camera.Since the blue line in the well target consists of a grid of 25 mm *25 mm and a red line grid of 5 mm *5 mm,the approximate position of the spot can be obtained by locating the blue-red grid area where the spot is located.Spot rebuilding coordinates can be obtained by converting the pixel coordinate system to the reconstructed coordinate system.Because the spot coordinates are known,the final calculation of the spot displacement generated by the laser emitter on the target is the tunnel surrounding rock deformation.In terms of algorithm,an improved adaptive median filter is used to eliminate the influence of tunnel poplar dust displayed by salt and pepper noise in view of the impact of dust on spot positioning during tunnel construction.For the influence of tunnel environment light,the blue line of target image is extracted by using the blue classifier based on YCr Cb color space,and the separated blue target is matched by pyramid layered template to realize the reconstruction of intersection space position of target,which is convenient for subsequent spot area positioning and precise positioning.The improved Otsu method is used to segment the spot image and extract the centroid coordinates of the spot.Finally,the blue-red grid area where the light spot is located is determined to complete the area positioning.For the spot located in the red grid area,the scale coefficient between millimeter and pixel is obtained by self-calibration,and the spot location is completed accurately.By calculating the distance between the coordinates of the spot before and after positioning,the shape variable of the monitoring point can be obtained.Experiments show that the system measurement error is 0.0662 mm.It conforms to the requirements of 0.5mm vertical and 3mm horizontal displacement stipulated in Survey Specification for Urban Rail Transit Engineering(GB/T 50308-2017).