Research on Statistical Algorithm of Medicine Boards Based on High Frequency Extraction and Contour Information

Aiming at the problems of low detection efficiency and high cost in the traditional counting process of medicine boards, a statistical algorithm of medicine plates based on high-frequency extraction and contour information is proposed. Firstly, It is necessary to preprocess the original image, then extract the high frequency part of the image and binarize it. After that, traverse each column of white pixels in the image and eliminate noise by filtering out. Next, use the Sobel operator to extract the longitudinal edges of the medicine boards and perform expansion processing. Finally, the method of contour detection is used to realize the counting of medicine plates. Through the test of the algorithm, the accuracy of the medicine plates counting can reach 96%, besides that, the detection time of each sample is less than 0.2S. The experimental results show that the method has high detection accuracy, which can achieve the real-time and rapid demand of the production line of pharmaceutical factories.