高精度实时激光加工路径识别研究

Path recognition research on real-time high precision laser machining

传统算法存在识别精度较低,实时性较差的问题。为了提高识别精度,提出基于视觉分析的高精度实时激光加工路径识别算法,首先利用合理的点运算分析法提高激光加工目标物体图像的对比度;其次通过迭代阈值分割法与二值图像闭运算提取激光加工目标物体图像特征;然后对提取的激光加工目标物体路径信息进行滤波处理;再利用适应度函数对激光加工目标物体路径进行排序;最后通过视觉分析法按排序结果进行依次识别,选取出激光加工目标物体的最优路径。实验结果分析证明,利用该算法能够实现激光加工目标物体路径的高精度识别,且实时性强。

The traditional algorithm has low recognition accuracy and poor real time performance. In order to improve these defects,this paper proposes a path recognition algorithm of real time high precision laser machining based on visual analysis. First of all,using reasonable point operation method to improve the contrast of laser processing target object image; Second,the image characteristics of laser processing targets are extracted by the iterative threshold segmentation method and the two-valued image closed operation. Then filter the extracted laser processing target path information. Sort the path by fitness function. Finally,the optimal path of laser machining target is selected by means of visual analysis. The experimental results show that the algorithm can achieve high precision recognition of object path of laser machining target.