Linear filtering of 2-D wavelet coefficients for denoising ultrasound medical images

The paper presents a novel denoising method for ultrasound medical images, whose quality is degraded by the peculiar phenomenon of speckle noise. The method is constructed step-by-step on the basis of recent research on the topic, and consists in Gaussian filtering of proper wavelet coefficients of the image, corresponding to vertical and diagonal details. A comparison with other filtering techniques for ultrasound imaging, i.e. Wiener and median filter, is presented. The obtained results, combined with those reported in independent research, demonstrate that the proposed denoising scheme has very good performance and is very promising for actual medical application.     

Autoregressive Wiener filtering in a scanning electron microscopy imaging system

In this paper, we propose to use the autoregressive (AR)-based interpolator with Wiener filter and apply the idea to scanning electron microscope (SEM) images. The concept for combining the AR-based interpolator with Wiener filtering comes from the essential requirement of Wiener filtering for accurate and consistent estimation of the power of the noise in images prior to filter implementation. The resultant filter is called AR-Wiener filter. The proposed filter is embedded onto the frame grabber card of the scanning electron microscope (SEM) for real-time image processing. Different images are captured using SEM and used to compare the performances of the conventional Wiener and the proposed AR-Wiener technique.     

一种用于自动调焦图像的降噪方法

图像清晰度评价是实现数字图像自动调焦的关键,而调焦图像降噪直接影响着图像评价的准确性。为有效抑制调焦图像中存在的两类主要噪声,结合椒盐噪声和高斯噪声特点,综合采用十字中值滤波法和维纳滤波法实现了图像降噪。实验结果表明,相比于单一降噪方法,综合方法不仅能最大程度地降低两类噪声,而且能较好地保护图像细节,可作为自动调焦成像系统图像预处理的主要方法。     

基于L-曲率流滤波器的图像降噪算法

提出了L-曲率流滤波器的图像降噪(滤波)算法,该方法按图像信噪比大小分高、中、低3类,分别由L滤波器降噪、多级L滤波器降噪以及多次迭代的组合滤波器降噪,并进行了实验研究。结果表明:该算法与均值和中值滤波器相比,输入图像信噪比越低,滤波效果越明显。当输入图像为低信噪比时,对于受高斯噪声污染的图像,该算法滤波比均值滤波平均提高2.98 dB;对于受脉冲噪声污染的图像,该算法滤波比中值滤波平均提高11.09 dB,说明该算法对降低不同种类和不同信噪比的图像噪声有较强的适应性。     

基于多级维纳滤波的OCT图像除噪方法研究

为了实现光学相干层析成像（OCT）过程中采集信号的噪声去除,设计了基于维纳滤波和对比度增强的OCT图像处理方法。首先对OCT图像的噪声进行了分析,继而针对OCT信号的噪声干扰问题提出一种除噪算法,通过多尺度维纳滤波器对图像进行滤波处理,然后根据区域特性采用对比度增强方法提高图像对比度。实验结果表明：经该方法处理后图像的背景方差（BV）较原始图像降低了3-7倍,而且信噪比高,细节方差和背景方差之比（DV/BV）提高了3-6倍。该方法不仅有效去除了OCT图像的噪声,而且视觉效果良好,是一种有效的图像后处理方法。     

基于脉冲耦合神经网络模型的小波自适应斑点噪声滤除算法

分析了维纳滤波原理和脉冲耦合神经网络(PCNN)模型的特点,根据斑点噪声统计模型的特征,结合小波变换方法,提出了一种基于PCNN模型的小波自适应斑点噪声滤除算法(W-PCNN-WD)来改善超声图像质量.首先,对超声图像进行对数变换,使斑点噪声转换为加性噪声;对医学图像进行维纳滤波处理,计算其加性噪声的标准方差,并以此作为小波阈值.然后,利用小波变换对图像进行预处理,利用PCNN在小波域中对小波系数进行相应的修正.最后,进行小波逆变换和指数变换,获得滤除噪声的图像.结果表明:本文提出的滤波方法优于其他滤波方法,当噪声方差为0.01时,本文滤波算法获得的峰值信噪比(PSNR)比经Wiener滤波方法获得的高出9 dB.该滤波方法能在有效去除超声斑点噪声的基础上保留图像的边缘细节信息,极大地改善了图像的视觉质量.     

一种有效的去除图像混合噪声的滤波算法

依据噪声特点以及图像的像素关联性,提出了一种有效的数字图像混合噪声滤波算法。该算法首先对图像像素进行检测,依据受脉冲噪声污染的图像像素与其周围邻域多数像素在亮度上具有显著差异的特点准确检测出脉冲噪声,然后以检测到的脉冲噪声为被处理像素,用其邻域内未被脉冲噪声污染的图像像素对其进行中值滤波,由于当前脉冲噪声点以及邻域脉冲噪声点均未参与滤波运算,从而较好地保护了图像的细节。对图像中的高斯噪声,提出了一种改进的加权均值滤波算法,该算法在定义相关度函数时,既考虑了像素的灰度相关性,又考虑了像素的位置相关性。实验结果表明:提出的混合噪声滤波算法不仅可以有效滤除图像中的混合噪声,而且还可以较好地保护图像的细节。     

运动模糊图像的实时恢复算法

提出了一种新的处理运动模糊图像的方法,以一维维纳滤波(1DWF)来快速恢复二维的模糊图像。给出了1DWF算法可行性的理论论证,并给出了1DWF算法及其它常用图像恢复算法的实验结果和对比评价。运算量分析表明:1DWF算法的运算量只有二维维纳滤波运算量的1/3;实验结果显示:在P4 3.0G主频的PC机上,1DWF算法可以使得512×512的8位灰度测试图像在36 ms内完成图像恢复。对图像恢复质量的评价结果显示:1DWF算法的图像恢复质量与原维纳滤波的恢复效果相同,验证了在单处理器上实时、高质恢复模糊图像的可行性。     

基于小波域维纳滤波的光纤面板暗影检测

通过分析光纤面板透光图像的噪声性质,提出在小波去噪的始末对图像取对数、指数变换,成功转换噪声模型,有效去除高斯及斑点噪声.并提出在小波域采用维纳滤波算法以增强去噪功能,实现更为有效地去除光纤面板透光图像中的噪声.最后对去噪图像进行暗影检测,实验结果表明,采用此算法进行去噪,检测出的暗影定位更加精准,冗余信息大量减少,有...     

离焦模糊图像的维纳滤波复原研究

微操作中，显微视觉系统获取的图像通常是离焦模糊图像。离焦模糊图像的退化模型可用圆盘函数描述，利用模糊图像无方向性的二阶拉氏微分图像的自相关的负相关峰形成的环形槽的直径等于作为圆盘函数直径的2倍可以确定该函数。对模糊图像进行一次维纳滤波方法得到原图像的估计值，然后利用该初始值求得原图像及噪声的谱密度估值，进而利用这些新获得的信息构成改进的维纳滤波器对退化图像进行第二次滤波。实验表明，该方法计算量小、鉴别精度高、抗噪声能力较强，突出原图像的一些关键细节，提高了图像的复原质量。     

无人机遥感山区公路图像去噪方法

为实现对遥感图像较好的滤波去噪效果并有效保持图像边缘细节,提出一种自适应开关均中值融合滤波方法,可有效处理无人机拍摄的公路遥感图像.首先,选定遥感图像的局部区域,并设定梯度变换阈值大小,同时为标记该区域内的各个像素点,定义标志数组.然后,与阈值进行比较,若当前用于比较的像素点的值比梯度阈值大,则进行自适应开关均中值融合滤波,利用自适应选定阈值来替代传统开关均中值滤波器的固定阈值;若当前像素点小于梯度阈值,则选取改进梯度倒数进行加权平滑处理,即针对图像局部区域建立相关函数,将权值参数按照局部统计特性适应性进行调整.由实验结果可知,与传统梯度倒数加权平滑算法相比,自适应开关内中值融合滤波算法能够滤除遥感图像中的脉冲和椒盐噪声,对其他类型的噪声也有较好的效果,同时保留了图像的边缘细节信息.实验结果表明,滤波处理的图像平均梯度指数提高了 ３．１６％ ,且参数均方误差( MSE )减少了约 ５％ ,具有较高的应用价值.     

Adaptive tuning piecewise cubic Hermite interpolation with Wiener filter in wavelet domain for scanning electron microscope images

Image processing is introduced to remove or reduce the noise and unwanted signal that deteriorate the quality of an image. Here, a single level two‐dimensional wavelet transform is applied to the image in order to obtain the wavelet transform sub‐band signal of an image. An estimation technique to predict the noise variance in an image is proposed, which is then fed into a Wiener filter to filter away the noise from the sub‐band of the image. The proposed filter is called adaptive tuning piecewise cubic Hermite interpolation with Wiener filter in the wavelet domain. The performance of this filter is compared with four existing filters: median filter, Gaussian smoothing filter, two level wavelet transform with Wiener filter and adaptive noise Wiener filter. Based on the results, the adaptive tuning piecewise cubic Hermite interpolation with Wiener filter in wavelet domain has better performance than the other four methods.     

盲复原高斯模糊图像

经典的图像恢复算法设点扩展函数(PSF)是已知的,然而在许多情况下PSF难以确定,不得不在只知道成像系统部分信息甚至没有任何信息的情况下估计真实图像和PSF,这一过程称为图像盲复原.对于高斯模糊图像,它的PSF是很难被检测出来的,因此高斯模糊图像的盲复原一直是个棘手的问题.利用高斯点扩展函数的特性,初始估计PSF并对加...     

Speech enhancement via Mel-scale Wiener filtering with a frequency-wise voice activity detector

This paper presents a speech enhancement system that enables a comfortable communication inside an automobile. A couple of novel concepts are proposed in an effort to improve two major building blocks in the existing speech enhancement systems: a voice activity detector (VAD) and a noise filtering algorithm. The proposed VAD classifies a given data frame as speech or noise at each frequency, enabling the frequency-wise updates of noise statistics and thereby improving the effectiveness of the noise filtering algorithms by providing more up-to-date noise statistics. The celebrated Wiener filter is adopted in this paper as the accompanying noise filtering algorithm, which results in significant noise suppression. Yet, the musical noise present in most Wiener filter-based systems prompts the idea of applying the Wiener filter in the Mel-scale in which the human auditory system responds to the external stimulation. It turns out that the Mel-scale Wiener filter creates some masking effects and thereby reduces musical noise significantly, leading to smooth transition between data frames.     

应用正则化影响函数扩散模型的星图噪声滤波

由于恒星星图的噪声滤波对保持星点的边缘细节要求较高,本文以塔基（Tukey）扩散模型与改善的PM（Perona-Malik）模型为基础,提出了一种基于正则化影响函数扩散模型的星图噪声滤波方法。该方法通过导数算子提取边界点集,利用图像中原始像素和噪声像素的空间分布特性对图像进行噪声滤波处理,并通过给定边界条件恢复图像边缘。由于避免了方差稳定（VS）变换,该方法可以直接处理高斯噪声。对普通图像和添加高斯噪声星图进行了仿真测试,并与普通扩散函数算法进行了比较。实验结果表明：提出的算法表现出了较好的噪声滤波能力,同时有效地保持了特征图像的边缘。相对于普通扩散函数算法其平均绝对误差降低了13.6%,峰值信噪比平均提高了6.1%。得到的数据显示,本方法的滤波能力优于普通的扩散函数方法,特别适用于星图的噪声滤波处理。     

航空多重模糊图像恢复算法

提出一种针对多重模糊的图像恢复算法,解决航空成像中多重模糊同时作用时的图像恢复问题。对多重模糊的物理模型进行分析,通过分步去模糊对多重模糊图像进行恢复。对空间不变模糊的点扩散函数进行合并,减少多重模糊恢复过程的计算误差累计,通过一次解卷积运算实现多重空间不变模糊图像的恢复。对包含空间变换模糊的多重模糊在分步恢复过程中所产生的图像误差噪声进行分析,使用维纳滤波对阶段误差噪声进行抑制,使得恢复图像的PSNR值提高7.76dB。实验结果表明基于点扩散函数合并的恢复方法能将多重模糊图像的PSNR值提高到28.09,有效地保障了图像的恢复质量。     

自适应窗口形状的中值滤波

中值滤波去除噪声的同时,难免出现含有线形图像的边缘轮廓信息跟随丢失的情况。为了达到滤除图像噪声最大化,有用信息损失最小化的目的,根据图像边缘特征选择或设计了适当大小和形状的窗口来进行中值滤波等操作。一方面,创新性地提出将Hough变换应用到车轮裂缝CT图像的滤波窗口形状的选择上,针对单一方向轮廓的图像,利用Hough变换检测出裂缝的方向,从而采用与裂缝形状相应的窗口对裂缝进行有针对性滤波,将该方法与传统方法进行了对比实验,并将60层滤波后切片图像堆栈得到了三维图像;另一方面,针对含有多个方向轮廓的图像,为了进一步改善滤波效果,提出根据像素梯度设计斜向滤波器,对多方向的线对进行滤波,数据显示经改进方法滤波后图像的峰值信噪比(PSNR)较传统的中值滤波提高了4~6,结构相似性(SSIM)提高了1%~2%左右,最后用683层荞麦切片CT图堆栈得到三维图像,对比滤波前后图像,可知该方法滤波效果良好。     

结合压缩感知和曲波的天文图像去噪

在天文图像去噪中,为了提高迭代曲波阈值算法的去噪重建性能,提出了基于循环平移和曲波维纳滤波的压缩感知迭代重构算法。首先,使用基于曲波阈值的循环平移方法对重构图像进行调整以抑制重构图像中的伪吉布斯效应;接着,用提出的曲波维纳滤波算子替代小波阈值在迭代过程中对图像曲波系数进行筛选以进一步提高重构图像的质量。通过对添加高斯白噪声的Lena图像和月球图像进行重构实验,分析本文算法和当前主流算法的性能。实验结果表明,与传统的压缩感知迭代曲波阈值算法相比,本文算法能够获得较优的去噪性能,有效地保护天文图像的细节信息,峰值信噪比大约提高了2.6~3.2dB。     

Robust evolutionary particle filter

The particle filter (PF) has been widely applied for non-linear filtering owing to its ability to carry multiple hypotheses relaxing the linearity and Gaussian assumptions. However, PF is inconsistent over time due to the loss of particle diversity caused mainly by the particle depletion in resampling step and incorrect a priori knowledge of process and measurement noise. To overcome these problems, in this paper, robust evolutionary particle filter is proposed. The proposed method can work in unknown statistical noise and does not require a prior knowledge about the system. In addition, to increase diversity, a resampling process is done based on the differential evolution (DE). The effectiveness of the proposed algorithm is demonstrated through Monte Carlo simulations. The simulation results demonstrate the effectiveness of the proposed method.     

Morphological gradient based adapted selective filter for removal of Rician noise from magnetic resonance images

Feature/edge-preserving noise removal techniques have a strong potential in several application domains including medical image processing. Magnetic resonance (MR) images have a tendency to gain Rician noise during acquisition. In this article, we have presented genetic algorithms based adapted selective non-local means (GASNLM) filter-based scheme for noise suppression of MR images while preserving the image features as much as possible. We have applied GASNLM filter with optimal parameter values for different frequency image regions to remove the noise. Filter parameter values are optimized by genetic algorithm (GA). A change in NLM filter known as selective weight matrix is also proposed to preserve the image features. The results prove soundness of the method. We have compared results with many well known and latest techniques, and the improvements are discussed.