The Research Of The Capsule Detection System Based On Image Processing

With the rapid development of modern technology and the progress of society, the quality detection requirements in all walks of life are constantly improved. the modern detection technology of supporting the automation production also are booming. Computer visual inspection as a nondestructive non-contact detection way to meet the detection requirements of modern industrial production, was given more and more attention. In the pharmaceutical industry, because the quality of pharmaceuticals related to the life safety, the detection requirements is more and more high.At present,in domestic,the capsule defects detection is completed artificially in general, this method is not only low efficiency and high missrate,but also the standards of artificial detection are different, and this non-contact detection easy cause secondary pollution, which cannot satisfy the requirement of modern industry. This paper developed a set of vision detection system to capsule defect based on image processing. This system is required to online detect various types of capsule defects fast and accurately, the detection speed is need not less than 75,000 grains per hour, capsule defects are mainly include ultra-long,ultra short, black spots, bubble, grease, wrinkles, crispy crack, different color, burr, etc. Also dimensional measurement,data storage and getting rid of rejected production automatically are required, The mainly research work in this paper are as follows:(1)Developed a set of vision detection system to capsule defect based on image processing, with a friendly man-machine interface, easy operation, with high stability and sophisticated function.System basically has the following four parts:image acquisition system, real-time display system, data processing system, control system.(2)Hardware design:the mechanical structure integrated design, the design of image collection system, the design of screener. In the limited conditions to design a set of stability, and feasible mechanical system to ensure the normal transmission, the sending trigger signals normally.the stability collecting of high quality images, screener runs normally.(3)Algorithm research:raise the filtering and enhancement algorithm for capsule image, to complete the extraction of capsule image, segmentation of image with uneven regions and extraction of defect feature.This research provide a set of specific detection algorithm to capsule defect, include related hardware system. It can be used to high-volume and fast detection, accurately recognize various defects, precisely get rid of rejected production. After debugging in the scene,test results prove it has very good stability,accuracy,practicality, is a cost-effective modern online detection equipment supporting our country medicine enterprise production equipment...