计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (1): 69-71.

• 学术探讨 • 上一篇    下一篇

自适应Bandelet框架及其在图像去噪中的应用

龙 奕1,2,尹忠科1,王建英1,李恒建1   

  1. 1.西南交通大学 信息科学与技术学院,成都 610031
    2.贵州大学 电气工程学院,贵阳 510003
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2008-01-01 发布日期:2008-01-01
  • 通讯作者: 龙 奕

Construction and application of adaptive bandelet frame in image denoising

LONG Yi1,2,YIN Zhong-ke1,WANG Jian-ying1,LI Heng-jian1   

  1. 1.School of Information Science & Technology,Southwest Jiaotong University,Chengdu 610031,China
    2.School of Electrical Engineering,Guizhou University,Guiyang 550003,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2008-01-01 Published:2008-01-01
  • Contact: LONG Yi

摘要: 在Bandelet变换中噪声影响了对图像实际几何方向的寻找。针对这一问题,提出了一种自适应Bandelet框架——根据图像去噪这一应用目标,重新修定建立四叉树结构和确定图像几何方向的若干规则,从而计算出较为精确的图像几何方向,并且实现了基于自适应Bandelet框架的去噪算法。实验表明同传统的小波子带多阈值去噪法相比,该算法不仅提高了去噪后图像的峰值信噪比(PSNR),而且更好地保留了图像的细节特征。

关键词: 自适应Bandelet框架, 四叉树, 几何方向, 小波阈值法

Abstract: The search for geometrical direction of image is counteracted by noise in Bandelet.Aiming at this deficiency,an adaptive Bandelet frame is presented;the rule of selection of best geometry of image and building of image quadtree is regulated afresh for the image-denoising objective in this frame.In accordance,the geometrical direction of image is calculated more exactly and a new image-denoising approach based on adaptive Bandelet frame is achieved.The numerical experiments show that this method gives better PSNR gains and can preserve much more details of the images than wavelet thresholding methods.

Key words: adaptive bandelet frame, quadtree, geometry, wavelet thresholding methods