机械手智能抓取工件时，工件抓取位置的获取是基于机器视觉精确装配的重点。针对环境较复杂的装箱管状工件的装配抓取环节，建立合理的机器视觉系统，利用正向照明系统对工件表面产生的反光直光条特征，提出了一种Gaussian拟合与Hough变换相结合的拟合算法。首先利用Gaussian拟合提取各光条法向上的中心坐标，然后将获取的坐标点集运用Hough变换进行拟合，最后根据获取的各光条中心线计算各工件的抓取位置。实验结果表明该方法能同时实现多条光条直线的拟合，又能抑制干扰点或噪声的影响，有效实现装箱工件抓取位置的智能获取。%When manipulator grasps intelligently the workpiece, the acquisition of grasping position about workpiece is the key of accurate assembly based on machine vision. For the grabbing link on assembly of packing tubular workpiece in the more complex environment as the object of study, a reasonable machine vision system is established, and the method combined with Gaussian fitting and Hough transform is proposed by using the feature of the forward lighting system reflecting straight light on the surface of workpiece. Firstly, Gaussian fitting method is used to extract the center coordinates of each light stripe in normal direction, then the coordinates of point set is fitted by using Hough transform, finally the workpiece grasping position is calculated according to the obtained each light center line. The experimental results show that this method not only can achieve the fitting of multiple light line, and can suppress interference and noise commendably, and effectively realize the intelligent acquisition of grasping position of packing workpiece.
展开▼
