Computer Engineering and Applications ›› 2010, Vol. 46 ›› Issue (27): 188-190.DOI: 10.3778/j.issn.1002-8331.2010.27.053

• 图形、图像、模式识别 • Previous Articles     Next Articles

Fingerprint image enhancement by directional diffusion PDE in wavelet domain

XIANG Ming,WU Xiao-pei,TANG Ting,SUN Lin-sen   

  1. Key Lab of Intelligent Computing and Signal Processing,Anhui University,Hefei 230039,China
  • Received:2009-02-25 Revised:2009-04-26 Online:2010-09-21 Published:2010-09-21
  • Contact: XIANG Ming


项 明,吴小培,汤 婷,孙林森   

  1. 安徽大学 计算智能与信号处理教育部重点实验室,合肥 230039
  • 通讯作者: 项 明

Abstract: A fingerprint image enhanced algorithm in wavelet domain based on PDE is proposed.The defects of coherent diffusion equation are discussed and a new nonlinear directional diffusion PDE is proposed to overcome its drawbacks.After wavelet decomposition,only approximation coefficients image is used to enhance the fingerprint image,the effect of the noise could be restrained.The experiment results on some images in FVC2002 DB4 database show that the proposed algorithm has better enhanced performance and lower computation burden than current PDE enhancement algorithm.

摘要: 提出了一种基于偏微分方程的小波域指纹增强算法。分析了一致性扩散方程的缺陷,改为使用方向扩散使方程严格沿着指纹纹路扩散;进而提出了一种基于方向扩散的小波域指纹增强算法。该算法利用小波域低频系数图增强指纹,抑制了噪声的影响,将增强子图利用小波逆变换实现重构。通过对FVC2002的dB4指纹库中部分低质量图像的增强结果比较,表明该算法对低质量指纹图像的增强效果明显,且处理速度比现存的基于PDE的增强方法快。

CLC Number: