基于多通道小波的DTI图像恢复

扩散张量图像中存在的赖斯噪声给张量计算和脑白质追踪等带来严重影响。为了减少噪声影响,该文采用多通道小波对扩散加权图像进行恢复,采用峰值信噪比来定量地评估本滤波器消除赖斯噪声的性能。基于模拟和真实数据对张量场的表面扩张系数等进行了计算并进行人脑白质纤维追踪。把该去噪方法和单通道小波方法进行比较,实验结果表明,提出的滤波器具有更好的噪声性能。     

DTI图像恢复的向量复扩散模型

为了减小扩散张量图像(DTI)中广泛存在的赖斯噪声影响,提出了向量复扩散模型.该模型是标量复扩散模型的推广和发展.为了评价该模型的去噪性能,对向量图像-扩散加权(DW)图像进行了恢复实验.基于模拟和真实数据进行的实验表明,相对于标量复扩散滤波器,向量复扩散滤波方法得到的PSNR和SMSE数值更高,追踪到的纤维数量更多、长度更长,故其去噪性能优于标量复扩散模型.另外,在信噪比较低情况下该模型优于实数域P&M向量滤波器.     

扩散张量成像去噪和纤维跟踪方法

提出一种纤维模型的数学合成方法，并利用合成模型对纤维跟踪方法进行了验证．对Streamline跟踪技术进行改进，采用能量最小化技术对面形或球形张量的跟踪方向进行校正以提高Streamline算法的准确性．为降低图像噪声对纤维跟踪的影响，分别采用小波去噪方法对扩散加权图像和张量场进行处理，并对小波去噪和传统的高斯平滑方法在扩散加权图像噪声抑制方面的作用进行了比较．     

复小波域维纳滤波与偏微分扩散相结合的图像去噪方法

针对图像噪声的去除,提出了一种基于复数小波域上的多方向窗维纳滤波与偏微分方程保持边缘细节相结合的方法。针对小波域维纳滤波的方向性差,去噪后图像容易产生哑铃效应,该方法首先进行双树复数小波变换,集中6个方向上的图像信号能量,之后,再在该6个方向上进行方向维纳滤波,对图像进行初步去噪,再以此引导偏微分方程中的扩散函数,实现各项异性进行扩散,最大限度地在保持图像细节的同时,去除噪声。实验结果表明,该方法的峰值信噪比,以及视觉质量都较复小波去噪或各项异性非线性扩散去噪方法有明显的改善。     

改进的各向异性复扩散模型的医学图像去噪方法

对医学图像进行有效的去噪并保持边缘信息,有利于图像的后续处理.本文分析P-M模型和Gilboa的复扩散模型以及它们的不足,提出一种改进的各向异性复扩散模型.该方法先用中值滤波对图像进行预处理,去除梯度值大的噪声点,然后用图像的虚部求扩散系数,以此引导扩散模型中的边缘检测函数,再进行八邻域像素的扩散过程.实验表明,该方法能达到较理想的去噪和保持边缘的效果,而且减少了迭代次数,缩短了计算时间.     

基于拓扑导数的复扩散在图像去噪及边缘提取中的应用

提出了基于拓扑导数的非线性复扩散用于图像去噪及边缘提取的一种算法.由于线性扩散会使图像边缘模糊,基于拓扑优化思想,对每个像素点的线性复扩散系数扰动,使得拓扑导数最小的扩散系数为最优.文中选取的扩散系数具有各向异性的特性,从而克服了Perona-Malik的各向同性扩散系数不利于去除边缘噪声的缺陷,选择拓扑导数足够小的像素点,对这些像素点用最优扩散系数进行扩散.文中给出了使算法迭代终止的判据.实验证明,与Guy Gilboa的非线性复扩散相比,本文方法对原始加噪图像处理后,实部图像体现出了更好的去噪效果,虚部图像则很好地保留了图像边缘,此外,本文方法还消除了Perona-Malik的方法对图像去噪后产生的阶梯效应.     

基于偏微分方程的舌图像去噪研究

在对舌图像的去噪过程中,平滑噪声的同时容易丢失边缘和纹理等细节信息。为此,研究基于偏微分方程的舌图像去噪方法,分别采用中值滤波、高斯滤波、P-M方程、正则化P-M方程以及耦合冲击-复扩散滤波模型,对加噪舌图像进行滤波。比较结果表明,正则化P-M方程更适合舌图像的去噪处理,该方法处理速度快、去噪效果好,且能有效保护图像边缘。     

结合非下采样轮廓波变换和混合阶次图像扩散的图像去噪

通过研究非下采样轮廓波变换理论及其在图像变换中的优点,提出一种新的基于非下采样轮廓波变换的图像去噪方法.该方法首先通过非下采样金字塔分解和非下采样方向滤波器组对待去噪图像进行非下采样轮廓波变换,然后采取不同阶次的图像扩散去噪算法分别对高频部分和低频部分进行去噪处理,最后将经过处理后的系数进行非下采样轮廓波逆变换便可得到去噪后的图像.通过实验结果表明,该方法不仅能有效的去除噪声,而且可以很好地保持边缘信息,整体性能优于近年来一些常见的去噪算法.     

一种改进的PDE图像去噪方法

研究用于图像去噪的偏微分方程;在理论上对去噪原理进行了分析。通过对扩散方程中扩散系数的改进,提出了一个对噪声图像更有效、更具有适应性的去噪扩散模型,对高斯噪声图像进行处理。与传统的各向异性扩散算法进行了比较并对偏微分方程的未来发展方向进行了展望。实验结果表明,该方法在有效去除噪声的同时较好地保留了图像中的重要细节信息,使图像的细节部分清晰。该方法可以有效地去除图像噪声,提高图像的质量。     

荧光免疫试纸条图像去噪方法分析

为削弱在荧光免疫试纸条检测系统中图像受到的污染,提出一种新的图像去噪方法。首先利用自适应中值滤波器消除对荧光图像影响严重的椒盐噪声,再用Contourlet变换图像阈值去噪方法对其余噪声进行处理。通过与其他去噪方法做对比,可知此方法在获得更好去噪效果的同时保护了图像轮廓细节。     

Seismic noise attenuation by means of an anisotropic non-linear diffusion filter

Noise attenuation is a major seismic data processing concern. In seismic data, noise can appear as random, coherent and/or impulsive. Recently, many different techniques, ranging from relatively simple processes to extremely complex ones, have been used for noise attenuation. Image filtering techniques are relatively new methods in seismic exploration. We introduced the anisotropic non-linear diffusion filter which is an effective way to de-noise images. Since a seismic section can be considered as an image of a two-variable function, we implemented the anisotropic non-linear diffusion filter to reduce both random and Gaussian noises. This filter is shown to be effective in removing noise while preserving edges and hence reducing resolution loss in seismic data. The anisotropic non-linear diffusion filter, with Tukey's function to guide the diffusivity, was applied to synthetic and real seismic data. The results show a signal-to-noise ratio increase with reflector continuity in addition to better recovery of reflector amplitudes even when dealing with complex subsurface geological structures.     

去除磁共振成像图像莱斯噪声的加权扩散

针对各向同性扩散易于造成图像边缘等特征区域的模糊以及相干增强扩散易于在图像背景区域内产生伪条纹的问题,提出了一种根据磁共振成像（MRI）图像莱斯噪声分布特点来对其进行降噪的加权扩散算法。该算法以MRI图像背景区域的莱斯噪声方差作为区分MRI图像背景区域和感兴趣的边缘特征区域二者特征差异的阈值。基于该阈值,该算法构造了一个加权函数,并用该函数对各向同性扩散和相干增强扩散进行加权。加权函数根据图像在不同结构区域的变化,自适应地调整两种扩散的权值,从而充分发挥两种扩散的优势并克服各自的不足。实验结果表明,该算法在峰值信噪比（PSNR）及平均结构相似度（MSSIM）的评价上优于一些经典算法。因此,该算法的降噪及保护、增强边缘的能力更为优越。     

金属断口图像去噪方法研究

分析金属断口图像可以为金属材料的性能及行为等许多方面的研究提供重要信息,准确分析金属断口的形貌需要对金属断口图像进行去噪等预处理。首先将图像划分为噪声、区域内部和边缘三类子图,然后针对受不同类型噪声污染的断口图像设计了自适应确定参数和中止条件的各向异性扩散的去噪方法,使滤波对不同噪声类型的适应性更好,对噪声的选择性平滑更准确,并能保护边缘和低对比度区域。实验结果表明该方法对噪声污染的金属断口图像有更好的抗噪效果和细节保持能力,有利于提高后续模式分析和定量计算的准确性。     

Rician noise removal from MR images using novel adapted selective non-local means filter

The reduction of rician noise from MR images without degradation of the underlying image features has attracted much attention and has a strong potential in several application domains including medical image processing. Interpretation of MR images is difficult due to their tendency to gain rician noise during acquisition. In this work, we proposed a novel selective non-local means algorithm for noise suppression of MR images while preserving the image features as much as possible. We have used morphological gradient operators that separate the image high frequency areas from smooth areas. Later, we have applied novel selective NLM filter with optimal parameter values for different frequency regions of image to remove the noise. A method of selective weight matrix is also proposed to preserve the image features against smoothing. The results of experimentation performed using proposed adapted selective filter prove the soundness of the method. We compared results with the results of many well known techniques presented in literature like NLM with optimized parameters, wavelet based de-noising and anisotropic diffusion filter and discussed the improvements achieved.     

核磁共振图像归一化互相关非局部自适应去噪

为了更好地去除核磁共振(MR)图像中莱斯(Rician)分布的噪声,首先提出使用图像局部归一化互相关(NCC)作为几何结构相似性的一个表征,对传统非局部算法中使用灰度计算像素相似性权值的方法进行有效补充;然后,将改进方法分别应用于非局部均值算法和非局部最小线性均方误差估计算法,并根据局部信噪比(SNR)动态自适应地计算非局部算法中待滤波像素自身的加权值或者像素之间相似性阈值,达到对核磁图像自适应降噪的目的。实验结果表明,该算法可以更好地抑制核磁图像中的莱斯噪声,有效保留图像中细节信息,对核磁共振图像进一步的分析研究以及应用于临床诊断等具有非常重要的应用价值。     

Determination of optimal parameters for bilateral filter in brain MR image denoising

Noise elimination is an important pre-processing step in magnetic resonance (MR) images for clinical purposes. In the present study, as an edge-preserving method, bilateral filter (BF) was used for Rician noise removal in MR images. The choice of BF parameters affects the performance of denoising. Therefore, as a novel approach, the parameters of BF were optimized using genetic algorithm (GA). First, the Rician noise with different variances (σ = 10, 20, 30) was added to simulated T1-weighted brain MR images. To find the optimum filter parameters, GA was applied to the noisy images in searching regions of window size [3 × 3, 5 × 5, 7 × 7, 11 × 11, and 21 × 21], spatial sigma [0.1–10] and intensity sigma [1–60]. The peak signal-to-noise ratio (PSNR) was adjusted as fitness value for optimization.After determination of optimal parameters, we investigated the results of proposed BF parameters with both the simulated and clinical MR images. In order to understand the importance of parameter selection in BF, we compared the results of denoising with proposed parameters and other previously used BFs using the quality metrics such as mean squared error (MSE), PSNR, signal-to-noise ratio (SNR) and structural similarity index metric (SSIM). The quality of the denoised images with the proposed parameters was validated using both visual inspection and quantitative metrics. The experimental results showed that the BF with parameters proposed by us showed a better performance than BF with other previously proposed parameters in both the preservation of edges and removal of different level of Rician noise from MR images. It can be concluded that the performance of BF for denoising is highly dependent on optimal parameter selection.     

Adaptive noise-reducing anisotropic diffusion filter

In image processing and computer vision, the denoising process is an important step before several processing tasks. This paper presents a new adaptive noise-reducing anisotropic diffusion (ANRAD) method to improve the image quality, which can be considered as a modified version of a speckle-reducing anisotropic diffusion (SRAD) filter. The SRAD works very well for monochrome images with speckle noise. However, in the case of images corrupted with other types of noise, it cannot provide optimal image quality due to the inaccurate noise model. The ANRAD method introduces an automatic RGB noise model estimator in a partial differential equation system similar to the SRAD diffusion, which estimates at each iteration an upper bound of the real noise level function by fitting a lower envelope to the standard deviations of pre-segment image variances. Compared to the conventional SRAD filter, the proposed filter has the advantage of being adapted to the color noise produced by today’s CCD digital camera. The simulation results show that the ANRAD filter can reduce the noise while preserving image edges and fine details very well. Also, it is favorably compared to the fast non-local means filter, showing an improvement in the quality of the restored image. A quantitative comparison measure is given by the parameters like the mean structural similarity index and the peak signal-to-noise ratio.     

On 1-Laplacian Elliptic Equations Modeling Magnetic Resonance Image Rician Denoising

Modeling magnitude magnetic resonance images (MRI) Rician denoising in a Bayesian or generalized Tikhonov framework using total variation (TV) leads naturally to the consideration of nonlinear elliptic equations. These involve the so called 1-Laplacian operator and special care is needed to properly formulate the problem. The Rician statistics of the data are introduced through a singular equation with a reaction term defined in terms of modified first-order Bessel functions. An existence theory is provided here together with other qualitative properties of the solutions. Remarkably, each positive global minimum of the associated functional is one of such solutions. Moreover, we directly solve this nonsmooth nonconvex minimization problem using a convergent Proximal Point Algorithm. Numerical results based on synthetic and real MRI demonstrate a better performance of the proposed method when compared to previous TV-based models for Rician denoising which regularize or convexify the problem. Finally, an application on real Diffusion Tensor Images, a strongly affected by Rician noise MRI modality, is presented and discussed.     

改进的P-M扩散方程在相干斑抑制中的应用

直接基于Perona-Malik扩散方程的滤波算法对于加性噪声非常有效,但是对于乘性噪声（如合成孔径雷达（SAR）图像相干斑噪声）收效甚微。提出了一种基于改进的Perona-Malik扩散方程抑制SAR图像相干斑噪声的新算法。分析对数变化对相干斑噪声的影响,为将P-M扩散方程应用于相干斑噪声抑制奠定了理论基础;通过P-M扩散和稳健统计学的联系,建立了基于Biweight Estimator误差模型的扩散系数;同时利用非线性衰减技术对梯度阈值的选择改进。实验表明,该方法不仅有效抑制了SAR图像相干斑噪声,较好地保持了细节和边缘信息,而且视觉效果比较好。