PE gas pipeline joint defect identification method and system

The invention discloses a PE gas pipeline joint defect identification method and system, and relates to the technical field of image identification, and the method comprises the following steps: obtaining a joint defect image of a to-be-identified PE gas pipeline; performing enhancement processing on the joint defect image by using a Laplacian algorithm; filtering and de-noising the enhanced joint defect image by using a median filtering algorithm; performing segmentation processing on the denoised joint defect image by using an adaptive mean threshold segmentation algorithm; performing edge detection on the segmented joint defect image, extracting the contour of the defect in the joint defect image and the contour between the joint defect image and the background, and converting the grayscale image into a binarized joint defect image; and constructing a convolutional neural network based on deep learning, and identifying the binarized joint defect image by using the trained convolutional neural network to ob