Aiming at the problem that it is difficult to detect the small defects with low contrast existing on the surface of lithium battery electrode plate, a new lithium battery electrode plate surface defect detection method based on improved Canny operator is proposed. Firstly, bilateral filtering is used to solve the image edge blur problem caused by Gaussian filtering when reducing noise, and on this basis, multi-scale detail enhancement algorithm is introduced to enhance the low contrast image. Secondly, the 3×3 gradient template of Sobel operator is used to calculate the gradient amplitude and gradient direction of electrode plate images. Finally, based on the maximum entropy and Otsu algorithm, the high and low thresholds of the image are automatically obtained. The detection results of the two algorithms after threshold segmentation are edge-fused with logic and operation, and the discontinuous edges are repaired with morphological closed operation and thinning algorithm to obtain the final detection edge. The experiment results show that the traditional Canny operator and Otsu-Canny algorithm are difficult to effectively detect the defects, including different types of dark spots, exposed foil and scratches. However, the algorithm in this paper can achieve good detection results for these defects, which can effectively reduce the same color background noise, while highlight the target defect area. The correct detection rate reaches 98% , the propose algorithm has a certain practical value.