Recognition of Similar Patterns by Mutilayer Nets and Detection of Rotated Angle and Scale Ratio
Hiromu Gotanda*, Kousaku Kawai** and Tatsuya Yamaoka***
*Department of Electrical and Computer Engineering, Faculty of Engineering, Kinki University, 11-6 Kayanomori, Iizuka-shi, Fukuoka 820-8555, Japan
**Sanki Medical Incorporation, 5-9-21 Nipponbashi, Naniwa-u, Osaka 556-0005, Japan
***Netmarks Incorporation, 1-3-12 Motoakasaka, Minato-ku, Tokyo 107-0051, Japan
Received:August 18, 1999Accepted:October 28, 1999Published:December 20, 1999
Keywords:mutilayer nets, BP learning, pattern recognition, rotated angle, scale detection
For practical pattern recognition, it is required not only to recognize geometrically similar patterns but also to detect the difference of translation, rotation and size from their templates. This paper proposes a method to recognize the simiar patterns by a mutilayer net and then detect the difference on the common basis of well-known geometrical characteristics (center of gravity, angle of principal axis, and variance). It is found from experimental resuts that, with the proposed method, a small net can cassify the similar patterns at a high recognition rate and detect their rotated angles and scale ratios with a high accuracy.
Cite this article as:H. Gotanda, K. Kawai, and T. Yamaoka, “Recognition of Similar Patterns by Mutilayer Nets and Detection of Rotated Angle and Scale Ratio,” J. Robot. Mechatron., Vol.11 No.6, pp. 495-501, 1999.Data files:
Copyright© 1999 by Fuji Technology Press Ltd. and Japan Society of Mechanical Engineers. All right reserved.