作　　者： (郭鹏跃）; (张嘉易）; (王晓蕊）; (郝永平）;

机构地区： 沈阳理工大学兵器科学与技术研究中心,辽宁沈阳110159

出　　处： 《仪表技术与传感器》 2017年第8期72-74,共3页

摘　　要： 针对微小零件的检测问题,开发了一套图像提取与特征点比对相结合的微小零件视觉检测系统。提出了用最大面积法去除轮廓噪声,获得所需的轮廓;提出了基于局部边缘最小二乘拟合线并求交点的特征点获取方法;并采用了实际图形与模板图形对比特征点的方法进行零件的检测,获得零件的误差信息。通过实验验证,所提出的方法能够较好地获取特征点,测量精度高。该方法适用于零件的自动检测,具有很好的鲁棒性。 For the problem of small parts detection,a small part visual inspection system combining image extraction and fea- ture point matching was developed.The maximum area method was proposed to remove the eontour noise,and the desired contour was ohtained.A feature point aequisition method based on local edge least squares fitting line and intersection point was proposed. And the actual graphics and template graphics feature points eomparision method of detection method was used to access to parts error message.The experimental results show that the proposed method can obtain feature points hetter and have accurate measure- ment aecuracy.This method is suitable for the automatic detection of parts,and has good robustness.

关 键 词： 轮廓拟合 图像处理 特征点 视觉测量 去噪 模板对比