Minimum spanning tree-based random-valued impulse noise detection for a switching median filter

A switching median filter is effective for impulse noise elimination while preserving edges and details of an image. In the switching median filter an impulse noise detector is employed before filtering, and the detection result is used to control whether a pixel should be filtered or not. However, the conventional impulse detector tends to misjudge noise-free pixels constructing line structures to be the noises. We propose a new random-valued impulse noise detector based on the minimum spanning tree, and it is applied to the switching median filtering to eliminate the impulse noise effectively even for the image including line structures. Through the experiments, the effectiveness of the proposed random-valued impulse noise detector is illustrated.     

Random-valued impulse noise removal by the adaptive switching median detectors and detail-preserving regularization

This paper firstly proposes an adaptive non-local switching median filter. Then, a two-phase scheme is presented to remove the random-valued impulse noise. In the first phase, the adaptive switching median filter or the adaptive non-local switching median filter is used to identify the pixels which are likely to be the noise candidates. In the second phase, only the noise candidates’ values are restored by a detail-preserving regularization method. Simulation results show that the proposed method is significantly superior to some of the state-of-the-art methods.     

Decision-based adaptive morphological filter for fixed-value impulse noise removal

A novel adaptive switching morphological filter for removing fixed-value impulse noise is proposed. The proposed filter firstly identifies noise pixels using the two-stage morphological noise detector, in which the initial noise detection is used to identify the noise candidates based on the morphological gradients and the refined noise detection based on the combined conditional morphological operators is adopted to further classify the noise candidates as the noise pixels or noise-free pixels. Then the detected noise pixels are removed by the adaptive morphological filter using the conditional rank-order morphological operators while the noise-free pixels are left unaltered. Extensive simulations show that the proposed filter outperforms a number of existing switching-based filters because of its excellent performance in terms of noise detection and image restoration.     

基于彩色数字图像处理的自动调焦技

针对彩色图像讨论了基于数字图像处理技术的自动调焦算法.根据彩色图像的特点，选取了能够表现图像清晰度的两种矢量范数——L1范数和L2范数分别作为粗略调焦和精确调焦的评价函数，以实现系统在大范围内的精确调焦.为了提高图像的质量，采用了既能保持图像细节又能滤除脉冲和高斯噪声的中值滤波算法和均值滤波算法相结合的矢量中值均值滤波器.实验结果表明，本调焦算法能够在大范围内调节焦距，且具有较高的调焦精度，调焦速度较快.     

闪光照相CCD图像的自适应中值滤波方法

中值滤波是一种在去除噪声的同时能较好地保护图像边缘细节的非线性图像处理方法。为了滤除闪光照相CCD图像中的脉冲噪声，同时能更好地保护图像边缘，提出了一种改进的自适应中值滤波方法。该方法采用局部中值和局部方差作为判断噪声点的阈值，实现了局部自适应的中值滤波，克服了传统自适应中值滤波方法的缺点，对椒盐噪声和随机脉冲噪声均有较好的滤波效果。实验结果表明，该方法消除图像脉冲噪声十分有效，对闪光照相CCD图像的处理结果也较好。     

基于改进开关中值滤波的多孔网栅图像脉冲噪声消除

针对多孔网栅闪光照相图像含有随机脉冲噪声的问题,提出了一种改进的开关中值滤波噪声消除算法。该算法利用像素与邻域窗口统计中值的灰度信息,建立噪声点探测器。通过设置噪声点探测阈值来识别噪声,并用邻域窗口内统计中值代替噪声点取值。经过多次滤波,含随机脉冲噪声的计算机合成网栅图像及实验网栅图像可获得良好的恢复效果。     

基于邻域特征匹配的通用噪声滤波器

图像去噪是遥感图像复原的重要步骤。在去除图像噪声的同时希望尽可能多地保留图像的纹理细节信息。受较差的成像环境和图像数据远距离传输的影响,遥感图像中一般都含有较强的高斯-脉冲混合噪声,而在现有的图像去噪算法中,能够同时去除图像中的高斯-脉冲混合噪声的通用噪声滤波器很少。以非局部平均方法的滤波思想为基础,通过引入邻域相似度评价的概念和脉冲噪声探测器,提出了基于邻域特征匹配的通用噪声滤波器。实验结果表明:基于邻域特征匹配的通用噪声滤波器具备有很好地去除图像高斯-脉冲混合噪声的能力,在去除高斯-脉冲混合噪声的同时能够很好地保持图像的复杂纹理和精细细节,并且便于向DSP/FPGA多处理器平台上移植。     

基于中值滤波和提升小波分析的图像去噪方法研究

针对现有算法大多对单一高斯噪声或脉冲噪声进行图像滤波的问题,在对二维图像平滑去噪的过程中,采用基于中值滤波和提升小波变换相结合的图像去噪方法。在中值滤波基础上,构造基于脉冲检测的中值滤波器,找出混合噪声中脉冲噪声并进行滤波;与此同时,对原始小波进行提升,构造提升小波,然后采用提升小波阈值去噪方法抑制高斯噪声。实验结果表明:采用本文方法,混合噪声得到有效抑制,去噪效果好。     

基于多引导滤波的图像增强算法

图像增强技术可以有效地突出图像中的有用信息,已广泛应用于多个领域.现有的图像增强算法往往无法应对自然图像中复杂的梯度分布,难以准确保持图像中前景与背景的边缘信息.为了改善输出图像的边界过平滑问题,本文提出了一个基于多引导滤波的图像增强算法.首先,设计了一个以滤波核为变量的通用图像优化模型,现有的联合滤波器可视为该模型的解;然后,依据集成学习的思想,将联合滤波器中的单幅引导图像扩展到多幅,以更好地利用引导图中的结构信息进而获得更好的输出结果,并给出了一个多幅引导图的来源途径;最后,对多幅输出图像进行平滑,在图像优化模型中加入正则化项,以确保由多引导滤波得到的不同滤波输出保持一致.实验结果表明,本文算法在抑制图像噪声的同时,可以更好地保留物体的边界信息,从而使图像的信噪比进一步提升.     

一种新的激光雷达距离像噪声抑制方法

主要针对激光雷达距离像的距离反常噪声抑制问题,阐述了激光雷达距离像的噪声原理,分析了应用传统中值滤波方法抑制距离反常噪声的缺陷,提出了基于包围准则的自适应中值滤波算法。该方法首先根据包围准则检测噪声,对5×5滤波窗口内的像素值进行排序差分;然后选择低于门限长度最长的连续差分值对应的像素值作为距离正常值;最后运用中值滤波和加权均值滤波进行噪声抑制。实验结果表明,该方法有效抑制了距离反常噪声,且较好地保护了距离图像中目标的边缘细节,均方根误差分别比3×3和5×5窗口中值滤波法减少了27.1%和9.1%。     

Morphological impulsive noise removal based on multi scale dual hit-or-miss transforms

Impulsive noise removal is an active research area in optical signal processing. Morphological operators have been tried for impulsive noise removal. However, the performance is not very effective. Dual hit-or-miss transforms could identify the protruding bright or dark pixels like impulsive noise pixels in image, which may be used to construct effective algorithm for impulsive noise removal. In this paper, an algorithm based on the multi scale dual hit-or-miss transforms is demonstrated. Firstly, the positive impulsive noise pixels, which are bright pixels, are identified by multi scale hit-or-miss transform. Then, the negative impulsive noise pixels, which are dark pixels, are identified by multi scale dual hit-or-miss transform. Finally, the identified impulsive noise pixels are removed and replaced by a reasonable value estimated by adaptive median filter. Experimental results show that, because the multi scale dual hit-or-miss transforms could effectively and correctly identify the impulsive noise pixels, the noise pixels are correctly removed and the real image details could be well maintained.     

自适应中值-加权均值混合滤波器

为了去除图像中混入的脉冲噪声和高斯噪声,提出了一种基于自适应中值滤波和自适应加权均值滤波的混合滤波方法。该方法先将图像分为若干区域,并对每个区域进行噪声检测以实现两类噪声的分离,然后再分别采用自适应中值滤波和自适应加权均值滤波将分离出的脉冲噪声和高斯噪声去除。对这种新方法进行了计算机模拟实验。结果表明:新方法较前人提及的三种混合滤波方法具有更优的滤波性能,在有效抑制混合噪声的同时能很好地保护图像中的细节,为消除图像中的混合噪声提供了一种有效的途径。     

基于改进中值滤波和提升小波变换的阈值去噪方法研究

对于实际拍摄的一些图像信噪比低,噪声密度大,且含有混合噪声,而现有算法大多只能去除单一噪声的问题。针对混合噪声中含有的脉冲噪声和高斯噪声,提出基于改进中值滤波和提升小波变换去噪相结合的方法。去噪过程中,使用中值滤波器提取脉冲噪声并采用中值滤波算法滤波后,构造提升小波,采用改进阈值函数提升小波阈值去噪方法去除高斯噪声。实验结果表明,当噪声值(,)=(0.4, 20)时,采用本文去噪方法,峰值信噪比（PSNR）为34.002 1,平均绝对误差（MAE）为2.365 3。     

Rician noise reduction in magnetic resonance images using adaptive non-local mean and guided image filtering

This paper proposes a Rician noise reduction method for magnetic resonance (MR) images. The proposed method is based on adaptive non-local mean and guided image filtering techniques. In the first phase, a guidance image is obtained from the noisy image through an adaptive non-local mean filter. Sobel operators are applied to compute the strength of edges which is further used to control the spread of the kernel in non-local mean filtering. In the second phase, the noisy and the guidance images are provided to the guided image filter as input to restore the noise-free image. The improved performance of the proposed method is investigated using the simulated and real data sets of MR images. Its performance is also compared with the previously proposed state-of-the art methods. Comparative analysis demonstrates the superiority of the proposed scheme over the existing approaches.     

Modifying NL-means to a universal filter

Despite a state-of-the-art filter for removing Gaussian noise, non-local means filter (NLM), like its local counterpart (the mean filter), is no longer so effective in removing salt-pepper noise which is common in real world as well. By contrast, adaptive median filter (AMF) is concise and can remove this type of noise effectively. Inspired by the AMF filtering strategies, in this paper, we modify NLM to a novel non-local universal filter (UNLM) which can remove not only either of Gaussian noise and salt-pepper noise but also their mixture. Experiments on artificial and benchmark images validate its feasibility and effectiveness.     

小波域高斯混合模型与中值滤波的混合图像去噪研究

基于高斯混合模型的小波去噪方法并结合中值滤波法对脉冲噪音有较好滤除效果的特点,将这两种方法结合起来,对含有高斯脉冲混合噪音图像进行去噪处理.该算法采用Matlab语言进行仿真.实验结果表明,这种混合去噪方法的效果要优于单纯使用中值滤波或者小波去噪的效果.     

辐射图像脉冲去噪方法

为去除辐射图像中的脉冲噪音,建立了一种先进行脉冲噪音检测再进行噪音数据修复的联合去噪音方法.首先采用测地膨胀方法从减灰度图像出发对原图像进行形态学灰度重构,再采用灰度阈值法从原图像与重构图像的差值图像中分割检测出脉冲噪音,然后采用改进非本地均值方法对噪音数据进行修复.改进非本地均值方法对相似像素的选择范围进行了限制,从原理上避免了传统非本地均值方法用于去除辐射图像脉冲噪音时易发生的以错纠错问题,本文对其进行了数学表述.实验证明,该方法对不同类型辐射图像的脉冲噪音均具有较好滤除效果,表现出良好的噪音数据修复和图像细节保护能力.     

Complex noise suppression using a sparse representation and 3D filtering of images

A novel method for the filtering of images corrupted by complex noise composed of randomly distributed impulses and additive Gaussian noise has been substantiated for the first time. The method consists of three main stages: the detection and filtering of pixels corrupted by impulsive noise, the subsequent image processing to suppress the additive noise based on 3D filtering and a sparse representation of signals in a basis of wavelets, and the concluding image processing procedure to clean the final image of the errors emerged at the previous stages. A physical interpretation of the filtering method under complex noise conditions is given. A filtering block diagram has been developed in accordance with the novel approach. Simulations of the novel image filtering method have shown an advantage of the proposed filtering scheme in terms of generally recognized criteria, such as the structural similarity index measure and the peak signal-to-noise ratio, and when visually comparing the filtered images.     

基于复小波和局部梯度的靶标图像混合降噪

提出了一种有效去除光电成像测量系统中靶标图像噪音的混合降噪法.根据图像像素局部梯度模找出图像中受椒盐噪音污染的像素,使用中值滤波降噪.对去除椒盐噪音的图像,利用复对数Gabor小波提取各像素的相位信息和幅度信息,确定最小尺度滤波器对噪音幅度分布的估计值,从而自动地确定各个尺度上的噪音幅度分布的估计值和噪音萎缩阈值,达到有效降噪的目的.实验表明,该方法的降噪效果明显优于实symlet4小波、中值滤波和单一复对数Gabor小波降噪法.     

非局部变分修复法去除高密度椒盐噪声

分析了中值滤波及其改进型算法在处理高密度椒盐噪声时效果不理想的原因,采用变分修复方法来去除高密度椒盐噪声,基于现有的全变差修复模型提出了非局部全变差修复模型。该模型利用椒盐噪声特点(均匀分布、灰度值为0或255),将噪声点看成是图像中遗失或是破损的点,首先在图像中寻找与噪声点邻域相似的区域,将相似区域的中心像素作为噪声点新的邻域然后对其插值,把图像降噪问题转化为图像修复问题,从而达到去除高密度噪声的目的。实验结果表明:该模型对噪声密度为90%的彩色和灰度图像去噪后,其峰值信噪比为22.85和28.77,在客观评价标准方面优于中值滤波及其改进型算法。该模型能有效去除高密度下的椒盐噪声并较好地恢复图像细节,为图像去除高密度噪声提供了一种新的途径。