Research On Crack Detection Technology Of High Pier By Wall Climbing Robot For Bridge

In recent years,bridge detection has been developing towards intelligence.Vehicle-mounted bridge crack detection system,non-contact detector,uav and wallclimbing robot are gradually used in bridge detection to overcome the disadvantages of traditional manual detection,such as poor safety,high cost,long detection cycle and sometimes traffic interruption.The wall-climbing robot has the characteristics of low safety risk,small size,light weight,low cost,high efficiency and strong mobility,etc.It can enter the blind area existing in traditional detection work for detection and complete the collection of diseases.At the same time,the digital image processing speed,high flexibility,high security advantages,to do further processing of the collected disease image,and then directly reflect the characteristics of the crack parameters,so as to achieve quantitative and qualitative analysis of the disease,complete the conventional bridge disease detection.In this paper,aiming at problems such as device stability of wall-climbing robot in detecting cracks on the surface of high pier concrete and extraction of image information of collected cracks,the technical route of numerical simulation and experimental verification is adopted to carry out research on the technology of wall-climbing robot for bridge in detecting cracks on high pier.The main research contents and conclusions are as follows:(1)participated in the research and development of the wall-climbing robot bridge detection device,and carried out load analysis,stiffness checking calculation and motion performance analysis,which proved that the device was stable and had good obstacle crossing ability under working conditions.Based on the results of finite element analysis,several optimization Suggestions are put forward.(2)according to the characteristics of the gray-scale value of crack image,genetic algorithm is applied to the threshold segmentation of crack image,and according to the gray-scale characteristics of crack image itself and the characteristics of the gray-scale value of background noise edge in the digital image matrix of crack,a matrix transform denoising method for crack image is proposed.For different crack images,different boundary values of crack and background gray scale can be obtained,and it can be used to solve the extraction of crack information in vertical crack,transverse crack and cross crack images(3)The eigenvalue extraction,fractal theory is used to crack fractal dimension is used to describe the actual length of the degree of the twists and turns to get cracks,through in different length and width of the precast concrete board reserved "cracks" design to further prove the applicability of the method,and on the basis of the theoretical derivation and fractal software proves that the method is also suitable for image processing caused after fracture cracks image.(4)Through the laboratory test,the device with camera completed the concrete wall crack image shooting,and after the shooting of the crack image to do further processing,extract the characteristic value of crack.The test proves that the intelligent detection device can effectively complete the detection of concrete wall cracks,and further proves that the crack image denoising method proposed in this paper and the fractal dimension crack eigenvalue acquisition method are also suitable for the crack image acquisition of the wall climbing detection device.