Objective: This study aimed to study the automatic detection method of common appearance defects in the production process of Qinqinchang. Methods: CMOS camera was used to obtain the image information of the product online, and image processing technology was used to extract the appearance defect features of the product and analyze the types of defects. Results: The accuracy rate of the online detection method for the common appearance defects of kissing intestines was 97.2%. Conclusion: The method is simple and has broad application prospect.