The paper aims to apply machine vision technology to detect the quality of commercial bar code to achieve integration, automation, intelligent inspection of EAN-13 commercial bar code. According to the requirements of detection, the acquisition device of commercial bar code image was designed and initialized by system calibration. For detecting bar code, its image was collected; the distortion was corrected soon afterwards. By edge detection, shape processing and contour matching, the exact position of bar code was found; the bar code image region was segmented, rotated and corrected, which was divided intobar code area and numeric area. For bar code area, the reflection curve was obtained by horizontal sampling; and the optical characteristic parameters and decoding data were calculated. Structural parameters were calculated combining the calibration parameters by analyzing the geometric characteristics and combining the calibration parameters. For numeric area, numerical recognition was accomplished by character segmentation and template matching. Finally, all detection data were obtained by comparison and analysis. 100 EAN-13 bar code samples were detected by this method and the traditional method respectively. Experiments showed that the test results of this method were in full compliance with national standard. The detection data were accurate and reliable. The method of this paper was operated quickly, the accuracy was higher, and the detection speed was greatly improved. In this paper, a new detection method is proposed, which realizes the integration, automation and intelligent processing of commercial bar code quality detection, and improves the inspection level and work efficiency significantly.