The quality of the terminal crimping surface has a great impact on the safety of new energy vehicle.Therefore,it is particularly important to improve the level of the terminal crimping surface quality analysis.Since the material of terminal and wire core mostly are copper,it is difficult to separate this two in terms of color on the terminal crimping surface.Therefore,image engineering technology is rarely applied to terminal crimping surface quality analysis.In this regard,using image processing technology,a terminal crimping surface quality analysis method based on segmentation is proposed.First,the foreground and background of terminal crimping surface photo are segmented through color space and Gaussian Mixture Model(GMM).Then,through the multiple Pyramid clustering algorithm,the wire core area and terminal area are divided.Finally,the thin copper wires in wire core are located and analyzed through morphological operations.This method solves the problem of difficult separation of terminals and wire cores with similar colors and textures,and achieves accurate positioning of thin copper wires in wire core.
color segmentationGauss mixture modelclustering algorithmmorphology
维普
万方数据
