| 基于局部保边函数的低信噪比图像去噪 |
| |
| 引用本文： | 何坤,周激流,刘昶.基于局部保边函数的低信噪比图像去噪[J].四川大学学报(工程科学版),2009,41(2):179-184. |
| |
| 作者姓名： | 何坤 周激流 刘昶 |
| |
| 作者单位： | 四川大学计算机学院,四川大学 计算机学院， 四川 成都 610065,四川大学 计算机学院， 四川 成都 610065 |
| |
| 摘 要： | 传统的去噪算法要求含噪图像信噪比较高，并且去噪后图像边缘及纹理信息受到不同程度地损失。本文针对传统算法的不足，提出了基于局部保边函数的低信噪比图像去噪算法，首先对低信噪比图像运用自适应中值滤波器减少椒盐噪声对图像的影响同时保留图像边缘和纹理等细节信息；其次分析处理后的图像局部邻域内像素之间的关系，设计图像局部保边映射函数，最后利用Poly-Ribière-Polak(PRP)算法求出目标函数的最值进而实现低信噪比图像的去噪处理，去除高斯噪声和残余的椒盐噪声。 与传统算法相比，本文去噪效果较好，尤其是对PSNR为5.4db的低信噪比图像去噪后图像PSNR 达到24.3dB。
|
| 关 键 词： | 去噪 低信噪比图像 局部保边函数 PRP算法 |
| 收稿时间： | 3/10/2008 4:15:48 PM |
| 修稿时间： | 8/1/2008 9:44:33 AM |
Low-SNR Image Noise Removal on Local Edge-Preserving Function |
| |
| HE Kun,ZHOU Ji-liu,LIU Chang,LIU Ran.Low-SNR Image Noise Removal on Local Edge-Preserving Function[J].Journal of Sichuan University (Engineering Science Edition),2009,41(2):179-184. |
| |
| Authors: | HE Kun ZHOU Ji-liu LIU Chang LIU Ran |
| |
| Affiliation: | computer college sichuan university,School of Computer Sci., Sichuan Univ., Chengdu 610065,China and School of Computer Sci., Sichuan Univ., Chengdu 610065,China |
| |
| Abstract: | Traditional noise removal arithmetic required that the signal-noise-ratio of the image is high, noise removed image by the traditional noise removal arithmetic loss large amounts of edge and texture information in the image. This paper presents a novel procedure for the low signal-noise-ratio (SNR) image noise removal on the local edge-preserving function which makes up the limitation of the traditional methods. Firstly, adaptive median filter is used to remove the part of salt-and-pepper noise and to preserve the edge and texture information in the image. Secondly, local edge-preserving function is built on basic of analyzing the relation of pixels in the local image block. Lastly, a minimization problem is solved by the Poly-Ribière-Polak (PRP) method in order to remove the Gaussian noise and the remnant salt-and-pepper noise in the image. Comparing the results of removing noise, noise removal efficient by our method is better. Especially when the PSNR of the image is 5.4db, the PSNR of the result image is 24.3db. |
| |
| Keywords: | Noise Removal low SNR image local edge-preserving function PRP method |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《四川大学学报(工程科学版)》浏览原始摘要信息 |
|
点击此处可从《四川大学学报(工程科学版)》下载全文 |
