基于机器视觉的某机座内孔表面缺陷在线检测系统设计

摘要

In order to improve the effect of compressor hole surface defect detection, a kind of test method was proposed in this paper, which is based on DM642 system and machine vision test methods. By this method, the surface image of the compressor hole through endoscope could be obtained, and then after image denoising and mean-shift segmentation processing in the DM642 system, the type and size of the surface defect can be identified. Finally the test results showed in the screen clearly. The workpiece code could be setted and the related hole surface image couled be saved through the keyboard.%针对某压缩机孔内表面缺陷检测问题,提出了一种基于DM642和机器视觉的检测方法.通过内窥镜获取机座孔内表面图像.在DM642中对图像进行去噪、mean shift分割等处理,识别出缺陷并计算缺陷大小.最后在显示屏上显示出检测结果.通过键盘可以设定工件编号和保存孔内表面图像.