基于稀疏表示与非局部均值的全球医学图像去噪方法

102 浏览量更新于2024-08-31 收藏 1.87MB PDF 举报

本文主要探讨了一种基于稀疏表示和非局部均值算法的全局医学图像去噪方法。该研究由复旦大学电子工程系的研究人员Yi Guo、Hanchao Chai和Yuanyuan Wang共同完成，并于2014年10月15日提交论文，12月20日接受发表。其工作被发表在《信息技术与生物工程》(IJBBB)杂志上，DOI为10.7763/ijbbb.2015.v5.391。研究的核心思想是利用过度完成的字典将医学图像分解成一个系数矩阵，其中大部分元素接近于零，这被称为稀疏表示。这种方法的优势在于，通过非局部均值算法处理这些稀疏系数，可以有效捕捉和利用图像中结构的重复性特征，从而实现真正意义上的全局去噪。这种方法不仅适用于超声图像，也适用于磁共振图像等不同医学成像模态。实验部分，作者通过合成和临床数据对这种新型去噪方法进行了对比测试，共选择了五种最先进的图像去噪技术进行评估。结果显示，基于稀疏表示和非局部均值的方法在改善图像质量和去除噪声方面表现出色，尤其是在处理细节保持和整体一致性方面，显示出明显的优势。这项研究的重要贡献在于它提供了一种创新且有效的去噪策略，对于提升医学图像的质量，减少误诊和提高医疗诊断准确性具有实际应用价值。它进一步证明了稀疏表示理论在图像处理领域的潜力，并为医学图像分析和计算机辅助诊断系统的优化提供了新的思考方向。

A Global Approach for Medical Image Denoising via Sparse

Representation

Yi Guo, Hanchao Chai

, Yuanyuan Wang

Department of Electronic Engineering, Fudan University, Shanghai, China.

* Corresponding author. Tel.: +86-21-65642756; email: yywang@fudan.edu.cn.

Manuscript submitted October 15, 2014; accepted December 20, 2014.

doi: 10.7763/ijbbb.2015.v5.391

Abstract: In this paper, a novel global noise reduction approach based on the sparse representation and

nonlocal means algorithm is proposed to enhance the image qualities of various medical imaging modalities,

including ultrasound images and magnetic resonance images. By using an overcomplete dictionary, a

medical image is decomposed into a sparsest coefficients matrix populated primarily with zeros. A nonlocal

means algorithm is developed to deal with these sparse coefficients to exploit the repetitive characters of

structures in the whole image, realizing a “truly” global denoising. With synthetic and clinical data of

ultrasound images and magnetic resonance images, this approach has been compared with other five

state-of-the-art denoising methods. The experiments quantitative results demonstrate the effectiveness of

our approach, especially superior in reducing the noise while well preserving the tissue details. It is

concluded that our proposed approach is capable of enhancing image quality in both ultrasound and

magnetic resonance images. It has a broad field of applications and will increase the diagnostic potential of

the medical images.

Key words: Global denoising, medical images, patches’ similarity, sparse representation.

1. Introduction

Due to the non-invasive, harmless and accurate nature, ultrasound imaging (US) and magnetic resonance

imaging (MRI) are widely used in diagnoses and treatments [1]. However, the imaging acquisition and

systems may introduce noises and artifacts, degrading contrast resolution and making medical images

difficult for non-specialists to interpret. Thus, to achieve the best diagnosis, it is essential to denoise medical

images without affecting important image features and destroying anatomical details [2].

In the past few decades, considerable efforts have been made in denoising the medical images. The local

statistics methods form an output by a weighted average calculation using sub-region statistics over

different pixel windows [3]-[5]. The median filter is a nonlinear operator, replacing the middle pixel in the

window with the median value of its neighbours, effectively in eliminating single and unrepresentative

pixels [6]. Anisotropic diffusion-based methods control the process of anisotropic diffusion based on

statistical characteristics of noise [7], [8]. The parametric estimation methods like the maximum-likelihood

filter and Bayesian denoising method establish noise models and adopt the robust parametric estimation

approaches [9], [10]. The wavelet-based denoising algorithms based on multi-scale decompositions of the

noisy images, apply soft thresholds to wavelet coefficients of different scales to eliminate the noise [11],

[12]. Although the above filters are capable of effectively suppressing the speckle pattern, they still remove

International Journal of Bioscience, Biochemistry and Bioinformatics

Volume 5, Number 1, January 2015

下载后可阅读完整内容，剩余9页未读，立即下载

weixin_38663197

粉丝: 8
资源: 926

基于稀疏表示与非局部均值的全球医学图像去噪方法

模糊神经网络像素分类的稀疏表示医学CT图像去噪方法.pdf

基于稀疏分解的图像去噪

matlab-基于Split-Bregman迭代算法的医学图像去噪matlab仿真-源码

稀疏表示理论在图像去噪中的应用：特征保留算法

3D变换域稀疏协作过滤：图像去噪新方法

稀疏表示与APBT在图像去噪中的探索：基于过完备字典与自适应学习

图像恢复技术：集中稀疏表示与稀疏去噪

基于Split-Bregman算法的医学图像去噪仿真研究

纹理引导的稀疏张量表示在肺CT图像去噪与增强中的应用

基于深度卷积神经网络的医学图像去噪研究

最新资源