计算机工程与应用 ›› 2014, Vol. 50 ›› Issue (6): 179-181.

• 图形图像处理 • 上一篇    下一篇

基于改进小波阈值函数的指纹图像去噪

黄玉昌,侯德文   

  1. 山东师范大学 信息科学与工程学院,济南 250014
  • 出版日期:2014-03-15 发布日期:2015-05-12

Fingerprint de-noising based on improved wavelet threshold function

HUANG Yuchang, HOU Dewen   

  1. School of Information Science & Engineering, Shandong Normal University, Jinan 250014, China
  • Online:2014-03-15 Published:2015-05-12

摘要: 针对指纹图像的特点,在分析了传统的软、硬阈值函数去噪原理的基础上,提出了一种新的小波阈值函数。新阈值函数克服了软、硬阈值函数存在的不足。通过选择不同的参数,可适应不同的图像。仿真实验表明,新阈值函数较传统阈值函数具有更好的视觉效果和更高的峰值信噪比(PSNR)。

关键词: 指纹图像, 小波变换, 新阈值函数, 图像去噪

Abstract: Focusing on the fingerprint’s features, a new wavelet threshold function based on the traditional soft and hard threshold function is presented. The new threshold function overcomes the weaknesses of soft and hard threshold functions. It can adapt to different images by selecting different parameters. Simulation results show that, compared with traditional threshold function, the new threshold function has better visual effects and PSNR.

Key words: fingerprint, wavelet transform, new threshold function, image de-noising