Feature Extraction of Seal Imprint Based on the Double-Density Dual-Tree DWT

  • Authors:

  • Li Runwu;Fang Zhijun;Wang Shengqian;Yang Shouyuan

  • Affiliations:

  • School of Information Technology, Jiangxi University of Finance & Economics, Nanchang, 330013, China;School of Information Technology, Jiangxi University of Finance & Economics, Nanchang, 330013, China;Jiangxi Science & Technology Teacher College, Nanchang, 330013, China;School of Information Technology, Jiangxi University of Finance & Economics, Nanchang, 330013, China

  • Venue:
  • ICCS '07 Proceedings of the 7th international conference on Computational Science, Part II
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

The most important problem on seal imprint verification is to extract the imprint feature, which is independent upon varying conditions. This paper proposes a new method to extract the feature of seal imprint using the double-density dual-tree DWT due to its good directional selectivity, approximate shift invariance and computational efficiency properties. 16 different directions information is obtained as a seal imprint image transforms by using double-density dual-tree DWT. Experimental results show that their directional behaviors are much different, although the frequency distributions of true seals are similar to false seals'. This method is stable and computationally efficient for seal imprint.