视觉机器人货物轮廓提取与定位

山东大学学报(工学版) ›› 2010, Vol. 40 ›› Issue (1): 15-18.

视觉机器人货物轮廓提取与定位

丛奎荣韩杰常发亮

山东大学控制科学与工程学院, 山东济南 250061

收稿日期:2009-04-21 出版日期:2010-02-16 发布日期:2009-04-21
作者简介:丛奎荣(1983-)，男，山东烟台人，硕士研究生，主要研究方向为模式识别与智能系统. E-mail:congkuirong@mail.sdu.edu.cn
基金资助:
国家自然科学基金资助项目(60775023，60975025)；山东省自然科学基金资助项目(Z2005G03)；山东省博士基金资助项目(2006BS01012)

Goods contour detection and positioning by a vision robot

School of Control Science and Engineering, Shandong University, Jinan 250061, China

Received:2009-04-21 Online:2010-02-16 Published:2009-04-21

摘要/Abstract

摘要：

提出一种视觉机器人货物拣选中的货物轮廓提取与定位方法.在图像预处理的基础上,采用Sobel算子提取货物与货箱边缘图像I₁［i,j］,通过图像投影取得货物支持区及其边缘图像I₂［i,j］,对货物支持区利用Canny算子提取边缘图像I₃［i,j］,利用I₃［i,j］对I₂［i,j］进行边缘连接,最后边缘跟踪提取货物轮廓实现货物定位,控制机械手运动并实现货物抓取.实验结果表明该方法实用有效.

关键词: 计算机视觉;边缘检测;轮廓提取;视觉机器人;货物拣选

Abstract:

A new method of goods contour detection and positioning by a vision robot was proposed. Based on image pre-processing and the Sobel operator, the edgeof goods and its container I_1［i,j］ was detected. The goods-supporting area and its edge image I2［i,j］ were calculated by image projection. Then anothergoodssupporting area’s edge image I3［i,j］ was obtained according to the Cannyoperatorand the breakpoints in I2［i,j］ were connected by comparing with I3［i,j］. Finally, the contour was extracted by the contour extraction method to locate the target and the robot was controlled to pick the goods. The experiments show that the method is effective and reliable.

Key words: computer vision; edge detection; contour extraction; vision robot; goods picking

丛奎荣韩杰常发亮. 视觉机器人货物轮廓提取与定位[J]. 山东大学学报(工学版), 2010, 40(1): 15-18.

CONG Kui-Rong, HAN Jie, CHANG Fa-Liang. Goods contour detection and positioning by a vision robot[J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2010, 40(1): 15-18.

参考文献

多维度评价

Viewed

Full text

Abstract

Cited

Shared

Discussed