The invention relates to a defect detection method based on a cyclic generative adversarial network and structural similarity, and the method comprises the following steps: S1, obtaining and preprocessing defect pictures, and enabling the defect pictures to serve as a training data set; s2, constructing a CycleGAN model, and training based on the training data set to obtain a model for mapping the defect picture into a defect-free picture; s3, inputting a to-be-detected image into the trained CycleGAN model, and comparing the difference between the original image and the repaired image by using a structural similarity algorithm to obtain a difference binary image; and S4, connected domain noise reduction and morphological processing are carried out on the difference binary image, and if the original image has defects, the white area in the binary image is the extracted defect shape. The method has the advantages of being high in detection precision, high in robustness to complex texture surfaces, capable of accurately detecting small defects and the like.