Design and research of injection products online detection image processing system

YU Xiao，XV Zhe-xiong

（Department of Electrical Engineering，Tongji University，Shanghai 201804，China）

Abstract：In order to simplify the precision instrument used in injection products detection，the image processing system used in defect detection was investigated. Based on the analysis of defect detection procedure，the concept of this image processing system was established. By comparing the difference between linear parameters in standard image and collected image of injection products，the amount of rotation could be got with the help of the image centroid. The amount of offset was measured by the parameter of line end point in Hough transform or correlation calculation based on FFT. Finally，whether the product was defective was judged by the regional

connectivity. Based on the simulation of Matlab and comparison between different images，the results demonstrate that，this program took into account the timeliness and accuracy of the system to meet the requirements of online detection of injection molding，and thus it can substitute the expensive mechanical structure of positioning devices。