基于物体颜色和尺寸特征的机器人视觉算法

谢春生; 郭帅; 程泓井; 李华伟

doi:10.13433/j.cnki.1003-8728.2014.1107

基于物体颜色和尺寸特征的机器人视觉算法

doi: 10.13433/j.cnki.1003-8728.2014.1107

1.
上海大学机电工程及自动化学, 上海 200444

详细信息

作者简介:
谢春生(1989- ),硕士研究生,研究方向为移动机器人和智能控制,xcscdv@163.com。

通讯作者:
郭帅,副教授,博士,guosliuai@shu.edu.cn.

计量
- 文章访问数: 189
- HTML全文浏览量: 48
- PDF下载量: 3
- 被引次数: 0
出版历程
- 收稿日期: 2013-03-20

A Robot Vision Algorithm Based on Object's Colour and Size Characteristics

1.
School of Mechatronic Engineering and Automation, Shanghai University, Shanghai 200444

摘要

摘要: 针对基于颜色特征的目标识别方法无法去除相同颜色干扰物的问题,提出了一种基于颜色和尺寸特征的目标识别算法.通过联合HSV(Hue,Saturation,Value)颜色空间转换,采用阈值法和种子生长法对特征颜色进行图像分割,通过双目视觉技术利用目标物体尺寸信息对分割后的图像区域进行过滤,获得颜色和尺寸均符合指定要求的物体图像区域,以此实现目标物体的识别,并提高物体识别的抗干扰能力.通过多次重复实验证明,在进行规则形状物体的识别时,该方法切实有效.
- 颜色 /
- 尺寸 /
- HSV颜色空间 /
- 机器人视觉
Abstract: In computer vision,there is a problem that the target recognition method only based on color feature can't filter the interference objects which have the same color with target object. A new target recognition method based on both object's colour and size features has been presented in this paper. In order to achieve the goal of object recognition,by combining the HSV colour space conversion,the threshold method and seed growth method have been used to implement image colour featrue extraction,and under the support of binocular vision technology,the size feature has been used to filter those image regions which have been extracted by image segmentation. The method has been proved by experiments to be effective in the regular shape object recognition.
- algorithms /
- binocular vision /
- colour /
- computer vision

HTML全文

参考文献(7)

[1]	杨其宇,彭刚,高键足球机器人系统中几种颜色模型的比较与分析[C]/马斯特杯2003年中国机器人大赛及研讨会,2003Yang Q Y, Peng G, Can J. Compare and analysis of somecolor models in MiroSot system[C]//Masters Cup 2003China Robot Competition and Seminars,2003(in Chinese)
[2]	许志倩,张作龙,朱万胜足球机器人视觉系统目标识别算法改进[J]流动传到与控制,2007,(3);1-3 Xu Z Q,Zhang Z L, Zhu W S. A target recognitionalgorithms improving of soecer robot vision system[J]Fluid Power Transmission and Control, 2007,(3):1-3(in Chinese)
[3]	马志伟,陈富林基于颜色的运动目标快速识别与跟踪[J]山西电子技术,2010,(1);28-30Ma ZW,Chen F L. Color-based moving targets quicklyidentify and track[J]Shanxi Elecaronic: Technology,2010,(1);28-30(in C hinese)
[4]	陈佳鑫,贾英民-种基于漫水填充法的实时彩色目标识别方法[J]计算机仿真,2012,29(3);4-9Chen J X, Jia Y M. Real-time color objecta recognitionmethod based on flood fill algorithm[J]ComputerSimulation, 2012, 29 (3);4-9(in Chinese)
[5]	杨淑莹VC++图像处理程序设计[M]北京:清华大学出版社,2003Yang S Y. VC++ image processing program design M]Beijing;Tsinghua University Press,2003(in Chinese)
[6]	贾得云机器人视觉[M]北京:科学出版社,2000Jia D Y. Robot vision[M]Beijing; Science Press,2000(in Chinese)
[7]	杨伟基于颜色学习的移动机器人目标跟踪研究[J]武夷学院学报,2011,30(5);73-78Yang W Objecta tracking of mobile robot based onstudying color [J]Journal of Wuyi University, 2011,30(5);73-78(in Chinese)