Identification of Typical Construction Defects of 10 kV Cable Intermediate Joints Based on SIFT Algorithm and Canny Algorithm

The uneven cutting of the main insulation and the outer semi-conductive layer and the burr of the pressure pipe are easy to lead to insulation failure. The image recognition technology can be used to intelligently control the construction defects. Aiming at the construction defects of cable intermediate joint, an edge detection method based on Canny operator is proposed. Firstly, the image of segmented cable joint is spliced by using Scale-in-Variant Feature Transform (SIFT) algorithm, and then the false edge can be removed and the edge information can be highlighted by using Canny operator, and the edge information of different parts of the intermediate joint can be obtained. The experimental results show that the cable detected by the algorithm in this paper can also accurately detect the defects of uneven peeling of semi-conductor and burr, and the accuracy of defect algorithm can reach more than 90%, which verifies the effectiveness of the algorithm.