江苏省科技期刊可视化阅读平台

导航条

江苏科技信息 ›› 2015, Vol. 32 ›› Issue (27): 43-45.doi: 10.3969/j.issn.1004-7530.2015.27.025

• 论文 • 上一篇    下一篇

机器视觉中平面靶标点自动识别方法研究

曹勇, 李凤娇   

  1. 南京赛宝工业技术研究院,江苏南京,210000;南京市计量监督检测院,江苏南京,210000
  • 出版日期:2015-09-25 发布日期:2015-09-25

Study on Automatic Method for Recognizing Feature Points of a Planar Target in Machine Vision

Cao Yong, Li Fengjiao   

  • Online:2015-09-25 Published:2015-09-25

摘要： 平面靶标已广泛应用于机器视觉领域,其中复杂现场环境下平面靶标点的自动识别具有较大难度.文章针对目前平面靶标点自动识别方法存在通用性差、易受异物遮挡影响等问题,提出了一种通用性强、不受背景及异物干扰影响的平面靶标点自动识别方法.该方法利用透视投影的交比不变性及靶标平面与其图像平面的映射关系,实现平面靶标点自动识别.实验结果表明:文章算法切实可行,鲁棒性强,可以实现复杂现场环境下平面靶标点的精确提取及自动识别.

关键词: 特征提取, 交比不变性, 自动识别

Abstract: Planar targets have been widely used in the field of machine vision ,and automatic recognition of the feature points on planar targets in complex site environment is of great difficulties. As the current methods of automatic recognizing feature points on planar targets are of poor robustness ,and are easily interfered by foreign objects,a novel method of automatic recognizing feature points,which is of high versatility and is not easily affected by background and foreign objects interferences ,is proposed in the paper. The cross ratio invariance and the homographic relationship between the target and the image plane are utilized in the method to realize automatic feature points recognition. Experimental results show that the method is viable. It can realize precise recognition of feature points on planar targets in complex site environment ,and it is of strong robustness.