Shape Recognition Based on Freeman Chain Code

In this paper, a shape recognition method based on Freeman chain code has been proposed. According to the detected corners, feature vector of a contour shape can be obtained. And the descriptor of each corner in the feature vector has three components, which are the normalized distance between the corner and the centroid of all the corners, the local main trend of both sides of the corners. By using the feature vector, the matching judgment has been given, and the results of the experiment are shown.

[1]  Bo Yu,et al.  A curve matching algorithm based on Freeman Chain Code , 2010, 2010 IEEE International Conference on Intelligent Computing and Intelligent Systems.

[2]  Vijay Laxmi,et al.  Corner Detection Using Difference Chain Code as Curvature , 2007, 2007 Third International IEEE Conference on Signal-Image Technologies and Internet-Based System.

[3]  Tae-Sun Choi,et al.  Shape Registration Based on Modified Chain Codes , 2003, APPT.

[4]  Cristina Urdiales,et al.  2D object recognition based on curvature functions obtained from local histograms of the contour chain code , 1999, Pattern Recognit. Lett..

[5]  Bo Yu,et al.  A corner detection algorithm based on the difference of FCC , 2010, 2010 International Conference On Computer Design and Applications.

[6]  Herbert Freeman,et al.  On the Encoding of Arbitrary Geometric Configurations , 1961, IRE Trans. Electron. Comput..