基于机器视觉的并联机器人工件识别定位

Location identification of parallel robot based on machine vision

  • 摘要: 针对并联机器人在分拣过程中无法识别特定工件,分拣精度低的问题,提出了一种基于机器视觉的工件识别定位技术。运用海康工业相机采集工件图像,对图像进行预处理,提出合适的方法提取图像边缘进而提取工件的轮廓;提取待识别图像中工件轮廓的Hu矩特征与工件模板轮廓的Hu矩特征进行相似性度量,识别工件。然后对其进行质心提取并确定工件位姿以对工件进行定位。实验结果表明,该方法能有效地对工件进行识别与定位,为工件识别提供了参考思路。

     

    Abstract: In order to solve the problem that parallel robot can not identify specific workpiece in the sorting process and the sorting accuracy is low, a workpiece recognition and positioning technology based on machine vision is proposed. The image of the workpiece is collected by the HK industrial camera, a suitable method is proposed to extract the image edge and then extract the contour of the workpiece. The Hu moment feature of the workpiece contour in the image to be recognized was extracted and the Hu moment feature of the workpiece template contour was extracted for similarity measurement to identify the workpiece. Then the center of mass is extracted and the pose of the workpiece is determined to locate the workpiece. Experimental results show that this method can identify and locate the workpiece effectively, and provides a reference for the identification of the workpiece.

     

/

返回文章
返回