Font Size: a A A

Research On Dripping Pills' Defect Detection And Sorting Device

Posted on:2016-02-01Degree:MasterType:Thesis
Country:ChinaCandidate:M LiFull Text:PDF
GTID:2322330479452644Subject:Mechanical and electrical engineering
Abstract/Summary:PDF Full Text Request
Dropping pill capsules are applied in food, tobacco and pharmaceutical industries. The defect sorting is conducted by the workers in the domestic. It is time-consuming and its quality cannot be guaranteed. The foreign sorting equipment are too expensive and difficult to get them. Moreover, the domestic equipment are only suitable for hard capsule and their performance cannot achieve the satisfactory level. Therefore, it's necessary to study the equipment used for detecting and sorting the defect dropping pills.Since the machine vision has widely used in the field of industrial detection and has obvious advantages, we adopt the machine vision technology to detect defects. In this paper, the defects' appearance features, namely chroma, grayscale and cylindricity, were used to sort the defected dropping pills. We studied the defect detection algorithm in MATLAB.We developed the detecting and sorting scheme based on machine vision, including the location and detection scheme, the image collection scheme, the transmission scheme and the sorting scheme. We adopted plexiglass turntable to deliver the dropping pills and electromagnet to sort out the defected dropping pills.The method of grouping staggered is proposed to balance detecting efficiency, transmission efficiency and sorting efficiency. The design of the plexiglass turntable and waste container is presented with detailed. The performance of the equipment, mainly included removing ratio and false detecting ratio, is verified by the experiment.It is concluded that the equipment can recognize the majority of defected dropping pills and sort them out. The work efficiency of the equipment is 180000 unit per hour. The removing ratio is 98% and the false detecting ratio is 2.2%.
Keywords/Search Tags:Dropping pill, Machine vision, Defect recognition, Structure design, Detecting and sorting, Experimental analysis
PDF Full Text Request
Related items