计算机科学 ›› 2013, Vol. 40 ›› Issue (Z11): 147-149.

• 智能控制与优化 • 上一篇    下一篇

一种改进的各向异性扩散去噪模型

赵海勇,贾仰理   

  1. 聊城大学计算机学院 聊城252000;聊城大学计算机学院 聊城252000
  • 出版日期:2018-11-16 发布日期:2018-11-16
  • 基金资助:
    本文受山东省自然科学基金项目(ZR2011FL023),山东省高校智能信息处理与网络安全重点实验室(聊城大学)资助

Improved Anisotropic Diffusion Denosing Model

ZHAO Hai-yong and JIA Yang-li   

  • Online:2018-11-16 Published:2018-11-16

摘要: 为了既能有效地去除噪声,又能够较好地保持图像的边缘以及重要的细节信息,在Perona和Malik提出的各向异性扩散模型(P-M模型)的基础上,通过对扩散方程中扩散函数的改进,提出了一种具备自适应性的去噪扩散模型,该模型对图像去噪处理更加高效。改进的扩散函数在梯度较小时为一个常数,大于某个阈值后变为单调递减函数,直至某个梯度时递减为零。以上扩散函数特性使各向异性扩散模型能够达到在同质区加速平滑、在边缘区停止平滑的目的。实验结果表明,改进的扩散模型是一种更为理想的保边缘平滑模型。

关键词: 图像处理,图像平滑,各向异性扩散模型,扩散函数

Abstract: In order to remove noise effectively and preserve key details,a more effective and adaptive diffusion denosing model was proposed.A new diffusion function was built based on the P-M model.The new diffusion function would be a constant when the gradient is low.After the gradient exceeds a certain threshold,the new diffusion function becomes a monotone descending function which reduces to zero at a certain gradient.The characteristic of the diffusion function makes the improved model smoothing faster in homogeneous area and stop in edge.The experiment results show that the performance of the improved model is better than the traditional methods.

Key words: Image processing,Image smoothing,Anisotropic diffusion model,Diffusion function

[1] 刘镇弢,李涛,杜慧敏,等.基于context模型的contourlet域图像去噪[J].计算机科学,2012,9(3):243-245
[2] 罗亮,冯象初,张选德,等.基于非局部双边随机投影低秩逼近图像去噪算法[J].电子与信息学报,2013,35(1):99-105
[3] Perona P,Malik J.Scale-space and edge detection using anisotropic diffusion [J].IEEE Transactions on Pattern Analysis and Machine Intelligence,1990,12(7):629-639
[4] Monteil J,Beghdadi A.A new interpretation of the nonlinear anisotropic diffusion for image enhancement [J].IEEE Transactions on Pattern Analysis and Machine Intelligence,1999,21(9):940-946
[5] 白俊奇,陈钱.基于各向异性扩散的红外图像噪声滤波算法[J].光学学报,2008,28(5):866-869
[6] Yu J H,Wang Y Y,Shen Y Z.Noise reduction and edge detection via kernel anisotropic diffusion [J].Pattern Recognition Letters,2008,29:1496-1503
[7] 王毅,张良培,李平湘.各向异性扩散平滑滤波的改进算法[J].中国图象图形学报,2006,1(2):210-216
[8] 林宙辰,石青云.一个能去噪和保持真实感的各向异性扩散方程[J].计算机学报,1999,1:1133-1137
[9] 赵德,何传江,陈强.结合局部熵的各向异性扩散模型[J].模式识别与人工智能,2012,5(4):642-647
[10] 王相海,李放.同向平均梯度的各向异性扩散模型[J].中国图象图形学报,2011,6(12):2105-2111
[11] Scott T A.Locally monotonic diffusion [J].IEEE Trans on Image Processing,2000,48(5):1379-1389
[12] 余瑞星,朱冰,吕梅柏.一种新的水平集停止项函数选取方法研究[J].系统仿真学报,2008,0(22):6154-6157
[13] 易子麟,尹东,胡安洲,等.基于非局部均值滤波的SAR图像去噪[J].电子与信息学报,2012,4(4):950-955
[14] 何艳敏,甘涛,陈武凡.基于稀疏表示的两级图象去噪[J].电子与信息学报,2012,34(9):2268-2272
[15] Papari G,Petkov N.Edge and line oriented contour detection:State of the art[J].Image and Vision Computing,2011,29(2):79-103
[16] Papari G,Petkov N.Adaptive pseudo dilation for Gestalt edge grouping and contour detection [J].IEEE Trans.Image Process,2008,17(10):1950-1962

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!