基于机器视觉的多类型工件测量系统研究

综合应用光电技术、数字图像处理技术和计算机技术,以薄片型机械零部件为主要测量对象,开发-套工件尺寸测量系统,主要包括照明系统、图像采集系统、图像处理系统和尺寸测量系统.根据检测要求,采用Canny算法进行边缘检测粗定位和8个方向模板的改进亚像素检测技术获得工件精确轮廓信息,再对轮廓进行分割和选择,采用Hough变换完成...

Research on Multi-type Workpiece Measurement System Based on Machine Vision

Taking thin mechanical parts as the main measurement object, a dimension measurement system for mechanical workpieces was developed by comprehensively applying photoelectric technology, digital image processing technology and computer technology, which mainly included lighting system, image acquisition system, image processing system and dimension measurement system. According to the detection requirements, Canny algorithm was used for edge detection and rough positioning, and improved sub-pixel detection technology with 8 direction templates was used to obtain the accurate contour information of workpieces, then the contour was segmented and selected. Hough transform was used to complete the two-dimensional dimension measurement of workpieces, and the two-dimensional dimension measurement system for multi-type workpieces was realized by using Halcon and C++ joint programming. The experimental results show that the average error between Hough transform measurement value and micrometer measurement value is 0.042 mm in linear and aperture measurement,the average time is 32.543 ms.