Strip steel surface defect detection based on machine vision is an important quality inspection method in the strip rolling process.In order to improve the efficiency and accuracy of strip surface defect detection, this paper proposes a new strip steel surface defect image edge detection algorithm.The algorithm first uses bilateral filtering to remove image noise to achieve the purpose of edge preservation and denoising, then uses an improved four-direction Sobel operator to detect the edges of defective images, and uses adaptive dynamic thresholds to select the best threshold for binarization.The valued image is processed by edge thinning based on Hilditch algorithm to obtain the final detection image.The algorithm is simulated on the Matlab platform, and the experimental results obtained are compared with the traditional Sobel operator. Experimental results show that the average segmentation accuracy of the algorithm in this paper reaches 93.5%.Compared with the traditional Sobel operator edge detection method, the algorithm in this paper can obtain better edge detection results.