The invention discloses a cloth defect detection method based on image processing. The method comprises steps as follows: a self-adaptive denoising algorithm is adopted according to the actual condition on site; (2) a denoised image is subjected to sharpening enhancement processing, so that texture details and edge contours of image defects are enhanced, and extraction of follow-up characteristicvalues is more reliable; (3) the image defects are subjected to segmentation, smoothing and circulation enhancement processing with a morphological algorithm and a circulation area marking method; (4)characteristic values of the cloth are extracted and normalized; (5) the cloth defects are recognized and classified. According to the cloth defect detection method based on image processing, an automatic cloth checking system based on image processing and information integration is researched and developed, and functions including automatic detection, finished product grading, quality statisticanalysis, information sharing, concentrated monitoring and management and the like of the cloth defects are completed, so that the common technical problems that the labor intensity is high, the production efficiency is low and the like when conventional manual cloth checking and manual counting are adopted in the textile dyeing and finishing industry are solved.