基于双目视觉的机器人运动目标检测与定位

针对移动机器人自定位精度低的问题,提出了先由静止的工作机器人进行自定位,再对运动目标进行检测和定位的方法.基于HSV模型颜色特征,工作机器人分割出人工路标并进行自定位,利用帧间差分法将采集到的视频图像序列中相邻两帧作差分运算,提取出运动目标,并通过双目立体视觉视差原理计算出运动目标的绝对坐标,帮助运动目标完成定位.结果表明,该方法定位精度高于传统的移动机器人自定位的定位精度,且算法的实时性好,具有现实的研究意义.

Moving object detection and positioning of robot based on binocular vision

In order to solve the problem that the self positioning precision of mobile robot is low, a method, where the self positioning of stationary work robot is firstly performed and then the moving object is detected and positioned, was proposed. Based on the HSV model colour feathers, the work robot divided the artificial landmark and performed the self positioning. In addition, the differential operation for two adjacent frames in the collected video image sequence was performed with frame differential method, and the moving object was extracted. The absolute coordinates of moving object were calculated with binocular stereo vision parallax principle, which could help the moving object to complete the positioning. The results show that the positioning accuracy of the proposed method is higher than that of traditional mobile robot, and the algorithm shows good real time performance and has the practical significance. 