Research on Underwater Surface Crack Detection Technology of Concrete Dam Based on Image Processing

The traditional method of manually detecting concrete surface cracks has the disadvantages of low efficiency and large error. Therefore, it is necessary to adopt non-contact underwater crack detection and identification method for concrete dam based on image processing technology. In this paper, the pixels of the image information collected by the system are calibrated applying MATLAB program. After image grayscale conversion, denoising and segmentation, the crack feature information in the image is extracted and its width is calculated. Finally, the identification accuracy of the detection technology is verified and the detection efficiency of underwater surface cracks of concrete dams is improved.