Research and Implementation of Partition Detection Algorithm Based on Defects of sanitary Products
In order to solve the problem of detecting the surface stains of the unfixed shading flexible sanitary Products,a new detection algorithm was proposed to realize the rapid and accurate detection of the unfixed shading flexible sanitary products.In the detection algorithm,the collected sample images are first pre-processed,and then the threshold image segmentation method is used in the sample images to perform partitioning,and different mathematical morphology processes are performed according to different partition regions;Then,the segmented image processed by mathematical morphology is subjected to defect detection of its suitable algorithm according to its different regional features;Finally,the stains are extracted by using the dilation,erosion,and connectivity of the region.The algorithm is implemented by Halcon operator of MVtec company in Germany.Experiments show that the algorithm can identify edge stains,cotton pad stains and cotton core stains well.The average time overhead is 92ms and the detection accuracy is 100%.Compared with the traditional method,the method in this paper is robust and fast,and can meet the needs of industrial high-speed production.
Shibin Chen Yingjie Tang Peng Zhao
School of Information Engineering,Beijing institute of Graphic Communication,Beijing 102600,China
国际会议
上海
英文
1-6
2018-12-17（万方平台首次上网日期，不代表论文的发表时间）