Identification method for number of magnet ring's coil based on mathematical morphology

LI Rongjia, HE Tongneng, GU Yuehan

(College of Information Engineering, Zhejiang University of Technology, Hangzhou 310023, China)

Abstract: Aiming at the existing problems when testing the products of winding machines, the image processing of winding products was achieved to identify whether the product is qualified. The image of magnet ring was analyzed by the program based on Matlab. Then, its number of coils was extracted. After reading the procedure, a pretreatment for those shoot images was realized, using the methods of format conversion, and grayscale conversion. By using image threshold segmentation, mathematical morphology processing and identification methods of counting, a data statistics for the automation testing was realized. A reference for automation product testing was provided. The exact number of coils as the products of winding machines needs to be tested and gualified. Through digital image processing simulation of Matlab, the theoretical stage work should be accomplished. The results indicate that the coils of winding products recognition rate reaches 99% on the Matlab.

Key words: winding machine; Matlab; number of coils; mathematical morphology; identification method