Seal Image Registration Based on Shape and Layout Characteristics

A novel seal registration method for Chinese seal verification is proposed in this paper using shape and layout characteristics. Seal registration is the process that aligns the sample seal imprint with the model seal imprint geometrically. The proposed method firstly classifies the seal imprints into one of the three shapes (rectangle, ellipse and circle) which are most common in China. Rotation angle of rectangular or elliptical seal imprint is estimated simultaneously with the shape classification procedure. Secondly, layout analysis and character distribution characteristics are used to estimate the rotation angle if the input is considered as a circular seal imprint. Finally, the input seal imprint image is rotated to horizontal direction and the best alignment between the sample seal image and the model seal image is made by translation transform. Experiments show that this proposed method performs very well in seal image registration. And the speed is much faster than the correlation- like methods proposed by other researchers.

[1]  Yung-Sheng Chen,et al.  Automatic identification for a Chinese seal image , 1996, Pattern Recognit..

[2]  Takahiko Horiuchi,et al.  Automatic seal verification by evaluating positive cost , 2001, Proceedings of Sixth International Conference on Document Analysis and Recognition.

[3]  Yung-Sheng Chen.  Computer processing on the identification of a Chinese seal image , 1995, Proceedings of 3rd International Conference on Document Analysis and Recognition.

[4]  Wen Gao,et al.  A system for automatic Chinese seal imprint verification , 1995, Proceedings of 3rd International Conference on Document Analysis and Recognition.

[5]  Takenobu Matsuura,et al.  Seal imprint verification with rotation invariance , 2004, The 2004 IEEE Asia-Pacific Conference on Circuits and Systems, 2004. Proceedings..

[6]  Katsuhiko Ueda,et al.  Automatic seal imprint verification system for bank check processing , 2005, Third International Conference on Information Technology and Applications (ICITA'05).

[7]  Katsuhiko Ueda.  Automatic Seal Imprint Verification System with Imprint Quality Assessment Function and Its Performance Evaluation , 1994 .

[8]  Qian Zhang,et al.  An automatic seal imprint verification approach , 1995, Pattern Recognit..

[9]  Takahiko Horiuchi,et al.  Automatic seal verification using three-dimensional reference seals , 1996, Proceedings of 13th International Conference on Pattern Recognition.

[10]  Katsuhiko Ueda,et al.  Automatic verification system for seal imprints on Japanese bankchecks , 1998, Proceedings. Fourteenth International Conference on Pattern Recognition (Cat. No.98EX170).