Quality Inspection of Necked-in Area in Pen Point based on Machine Vision
Liping Tang, Fen Chen, Hengchang Guo · Advances in engineering research/Advances in Engineering Research · 2017
In order to solve the low efficiency and low accuracy of manual inspection in the quality of necked-in area in pen point, a program of pen point quality inspection was developed based on machine vision in Halcon platform.Firstly, use the median filter to remove the image noise and enhance the contrast of the image; then get the object region of the image through Blob analysis and use edge detection to extract the sub-pixel edges of the object region; finally, use the least square method to fit the edges, and detect the defects by the difference image algorithm.Experimental results show that the program can measure the thickness of the necked-in area quickly and accurately, and identify the defect efficiently.