Packaging Bag Edge Detection Based on Improved Canny Algorithm

Aiming at the problem that the gray contrast of packaging bag patterns is not obvious in edge detection, an edge detection method of packaging bags based on improved Canny algorithm is proposed. First, the image is grayscale stretched. Second, the threshold is determined adaptively using Augmented Otsu. Enhanced Otsu adaptation makes the edge information of the detected image richer and the overall information more complete. Third, the obtained threshold is input into the Canny algorithm to obtain the edge detection result. Finally, the experimental results are analyzed and the information entropy is calculated. The results show that the edge image output by the improved Canny algorithm is richer in edge information, and the information entropy is increased by 12.63%, 5.09%, 3.18%, and 5.41%, respectively.