Computer Engineering and Applications ›› 2008, Vol. 44 ›› Issue (21): 162-165.DOI: 10.3778/j.issn.1002-8331.2008.21.045

• 机器学习 • Previous Articles     Next Articles

Algorithm based on texture character analysis for fingerprint image segmentation

ZHAN Xiao-si,MENG Xiang-xu,YIN Yi-long   

  1. School of Computer Science & Technology,Shandong University,Ji’nan 250101,China
  • Received:2008-04-30 Revised:2008-06-10 Online:2008-07-21 Published:2008-07-21
  • Contact: ZHAN Xiao-si

基于纹理分析的指纹图像分割算法

詹小四,孟祥旭,尹义龙   

  1. 山东大学 计算机科学与技术学院,济南 250101
  • 通讯作者: 詹小四

Abstract: It is one of the important research emphases to process the fingerprint image with low-quality.Segmenting the fingerprint image with low-quality exactly and quickly is an important step for improving the performance of the automatic fingerprint recognition system.After analyzing the limitation of the fingerprint image segmentation algorithm based on gray variance,the paper introduces the texture character as the segmentation index,analyzes the distributing rule of the pixels in the fingerprint image and proposes the fingerprint image segmentation algorithm based on the texture analysis.Experimental results indicate that the algorithm has higher segmentation performance than the fingerprint image segmentation algorithm based on the gray variance.And it is robust for the fingerprint images whose background is more complicated.

Key words: fingerprint, fingerprint recognition, fingerprint segmentation, texture

摘要: 低质量指纹图像处理是近年来自动指纹识别技术的研究重点,对低质量指纹图像的分割是实现后续处理的前提。文中在分析了方差作为分割指标的局限性基础上,从指纹图像的纹理特征出发,研究了指纹图像的灰度分布规律,提出了基于纹理的指纹图像分割算法。实验结果表明,相比于基于灰度方差的指纹图像分割算法,文中算法的分割效果更好,对噪声的抵抗能力更强。

关键词: 指纹, 指纹识别, 指纹分割, 纹理