基于ROI的银触点模板匹配缺陷的检测法

A Detection Method for Matching Defects of Silver Contact Template Based on ROI

银触点广泛应用于各种电子元器件中,其表面要求覆盖银层完整光亮,但在生产中表面会出现多种缺陷难以检测而影响使用。为保证其出厂的零缺陷,提出了基于ROI的银触点模板匹配缺陷检测方法。通过对图像进行形态学处理,保证了整张图像的低噪声,根据银触点的表面外形特征,优化了模板匹配的搜索路径,实现了自动选取目标。凸包运算后准确做出圆形区域的最小外接矩形ROI,通过位址映射求出最小外接矩形的位置,得到信息完整的原图ROI。将模板ROI与实时ROI进行去均值归一化相关匹配,计算两幅图像的相似度。通过大量实验得出合格约束相似度,对比人工分拣与机器视觉ROI模板匹配的分拣误差,得出基于ROI的银触点模板匹配缺陷检测方法,符合银触点的出厂要求。

Silver contacts are widely used in a variety of electronic components.Its surface requires the contact is covered with a full bright silver layer.Whereas in production,the surface will appear a variety of defects which may affect the further uses.To ensure the zero defect of the factory,based on the principle of machine vision,a method of silver contact template matching defect detection based on ROI was proposed.Morphological processing of the image ensured the low noise of the whole image.The search path of template matching was optimized according to the surface characteristics of silver contact,implementing automatically select targets.The circular area of the smallest circumscribed rectangle for the ROI was foundwith optimization convex hull.The position of the minimum circumscribed rectangle was obtained by the address mapping,then the originalimage ROI with information integrity was obtained.The full-search zero-mean normalized cross-correlation algorithm(ZNCC)matched the template ROI with the real-time ROI.The similarity between two images was calculated.The similarity degree of qualified constraint was obtained by a large number of experiments.Contrasted with manual sorting,the similarity of two images was calculated.Based on sorting errors from template matchingin the machine vision ROI system,it was concluded A detection method for matching defects of silver contact template based on ROI which conforms to the factory requirements of silver contact.