无序安全扣快速识别和定位方法设计

Design of rapid identification and positioning of disordered safety buckles

摘要:

针对传统的安全扣生产在存在位置摆放存在偏差时，无法实现夹取的问题，提出了一种无序安全扣快速识别和定位方法。该方法采用一种基于子模板和图像金字塔的改进NCC算法进行模板匹配，并结合圆投影算法实现对旋转安全扣的目标检测，同时采用基于交叉匹配和RANSAC方法进行安全扣精确定位。实验验证通过将定位结果与实际坐标进行对比，结果表明识别误差为0~1个像素，其中X轴的平均误差仅为1.02mm，而Y轴的平均误差为0.76mm，实现了无序安全扣精确地识别和定位，能够满足工程上的实际需求。

Abstract:

Aiming at the traditional production of safety buckles that cannot achieve clamping when there is a deviation in position placement, a method for rapid identification and positioning of disordered safety buckles is proposed. The method adopts an improved NCC algorithm based on sub-templates and image pyramids for template matching, and combines with the circular projection algorithm to achieve the target detection of rotating safety buckles, and also adopts the cross-matching and RANSAC based methods for accurate positioning of safety buckles. The experimental validation is carried out by comparing the positioning results with the actual coordinates, and the results show that the recognition error is 0~1 pixel, in which the average error of the X-axis is only 1.02mm, while the average error of the Y-axis is 0.76mm, which achieves the accurate recognition and positioning of the disordered safety buckles, and is able to satisfy the practical needs in engineering.