Vision based Insert Defect Detection Method using Shape Symmetry and Image Similarity

摘要

The present invention relates to a method of detecting a defect of a cemented carbide insert by obtaining images of the cemented carbide insert through a camera of high resolution and comparing the symmetry of the product and the image similarity based on the image to prepare a cemented carbide insert step; Obtaining an optical processing and image of the cemented carbide insert; Estimating a shape of the sample through a mathematical analysis based on the image acquired in the step and classifying the region according to the symmetry; Obtaining a partial image of an inspection region to detect a defect based on the image acquired in the step and a corresponding region classified in the step; And detecting a defect by comparing the similarities between the partial images obtained in the step. The present invention relates to a method of detecting a defect of a cemented carbide insert using image symmetry and image similarity. Therefore, the defects can be detected by comparing the similarity with each corresponding image by using the symmetry of the product to check defects having various shapes and sizes.