中国邮电高校学报(英文) ›› 2010, Vol. 17 ›› Issue (6): 101-105.doi: 10.1016/S1005-8885(09)60532-X

• Others • 上一篇    下一篇

Image corner detection using topology learning

孙伟1,杨烜2   

  1. College of Computer Science and Software Engineering, Shenzhen University, Shenzhen 518060, China
  • 收稿日期:2010-03-05 修回日期:2010-07-15 出版日期:2010-12-30 发布日期:2010-12-06
  • 通讯作者: 孙伟 E-mail:sunweidemail@gmail.com
  • 基金资助:

    国家级.国家自然科学基金

Image corner detection using topology learning

  1. College of Computer Science and Software Engineering, Shenzhen University, Shenzhen 518060, China
  • Received:2010-03-05 Revised:2010-07-15 Online:2010-12-30 Published:2010-12-06
  • Contact: SUN Wei E-mail:sunweidemail@gmail.com

摘要:

Image corner detection plays an important role in image analysis and recognition. This paper presents a novel corner detector based on the growing neural gas (GNG) network and this proposed detector is called GNG-C. With the GNG network, image topology information can be learned and used to implement corner detection. The GNG-C approach can be described as consisting of the following steps. First, a canny edge detector is used to acquire the contour information of the input image. This edge information is used to train a modified GNG network. A special stopping criterion is defined to terminate network learning. Second, vectors formed between network nodes and their neighbors are used to measure curvatures. Third, dynamic regions of support (ROS) are determined based on these curvatures. These ROS are used to suppress curvature noise. The curvature values of the nodes are then analyzed to estimate the candidate corners. Finally, the candidates are distilled by a non-maxima suppression process to obtain the final set of corners. Experiments on both artificial and real images show that the proposed corner detection method is feasible and effective.

关键词:

corner detection, GNG topology learning, self-organizing map

Abstract:

Image corner detection plays an important role in image analysis and recognition. This paper presents a novel corner detector based on the growing neural gas (GNG) network and this proposed detector is called GNG-C. With the GNG network, image topology information can be learned and used to implement corner detection. The GNG-C approach can be described as consisting of the following steps. First, a canny edge detector is used to acquire the contour information of the input image. This edge information is used to train a modified GNG network. A special stopping criterion is defined to terminate network learning. Second, vectors formed between network nodes and their neighbors are used to measure curvatures. Third, dynamic regions of support (ROS) are determined based on these curvatures. These ROS are used to suppress curvature noise. The curvature values of the nodes are then analyzed to estimate the candidate corners. Finally, the candidates are distilled by a non-maxima suppression process to obtain the final set of corners. Experiments on both artificial and real images show that the proposed corner detection method is feasible and effective.

Key words:

corner detection, GNG topology learning, self-organizing map