药柱银丝位置多重算法识别技术研究

Research on the Recognition of the Position of the Silver Wire of Grain Based on Multiple Algorithms

提出了一种图像特征点筛选提取的方法对药柱中银丝位置进行快速有效的精确识别定位,该方法在对CCD相机获取的药柱端面图像进行二值化预处理的基础上,通过边缘检测与椭圆拟合结合的方式,提取环形ROI区域,并利用标记法比对ROI区域内各特征区域像素个数,排除干扰区域,最后采用贪心算法,比对剩余特征区域的中心坐标之间距离,最终得到银丝位置坐标;该方法有效的排除了药柱端面的干扰特征,实现快速准确的提取符合银丝位置的特征区域,满足批量生产的要求。

A method of image featmure point selection and extraction is proposed, so that the position of silver witre in the grain can be quickly and accurately identified and located. Based on the binarization preprocessing of the image obtained by CCD camera, the circular ROI region is extracted by combining edge detection with ellipse fitting method. Finally, the greedy algorithm is used to compare the distance between the center coordinates of the remaining feature regions, and finally the position coordinates of silver wire can be obtained by comparing the number of pixels in each feature region in the ROI region and eliminating the interference area. The results show that the method can effectively eliminate the interference features on the end surface of the grain, and realize the rapid and accurate extraction of the characteristic region in line with the position of silver witre, and meet the requirements of mass production.