Median-type filters with model-based preselection masks

To remove impulsive noise, several variants of median-type filters are presented. They use some prior information provided by the probabilistic composition model of the class of the image being processed. This information is doubly applied; first, in elaborating a noise-estimation mask for the preselection of the pixels which must be filtered; and second, in the filtering process itself. The aim is to avoid the side-effect damage that median-type filters produce in processing the non-contaminated pixels. Experimental results are better than in the conventional median filter.     

Pipelined robust M-estimate adaptive second-order Volterra filter against impulsive noise

The main limits on adaptive Volterra filters are their computational complexity in practical implementation and significant performance degradation under the impulsive noise environment. In this paper, a low-complexity pipelined robust M-estimate second-order Volterra (PRMSOV) filter is proposed to reduce the computational burdens of the Volterra filter and enhance the robustness against impulsive noise. The PRMSOV filter consists of a number of extended second-order Volterra (SOV) modules without feedback input cascaded in a chained form. To apply to the modular architecture, the modified normalized least mean M-estimate (NLMM) algorithms are derived to suppress the effect of impulsive noise on the nonlinear and linear combiner subsections, respectively. Since the SOV-NLMM modules in the PRMSOV can operate simultaneously in a pipelined parallelism fashion, they can give a significant improvement of computational efficiency and robustness against impulsive noise. The stability and convergence on nonlinear and linear combiner subsections are also analyzed with the contaminated Gaussian (CG) noise model. Simulations on nonlinear system identification and speech prediction show the proposed PRMSOV filter has better performance than the conventional SOV filter and joint process pipelined SOV (JPPSOV) filter under impulsive noise environment. The initial convergence, steady-state error, robustness and computational complexity are also better than the SOV and JPPSOV filters.     

A state variable approach to digital image processing

Digital filtering of images is considered. A recursive filter design for two-dimensional separable filters is discussed which is based on state variable methods. Advantages of this design are guaranteed stability, and shorter word length than filters designed by bilinear mapping techniques, greater computational efficiency and easier design as compared with nonrecursive filters. Results are shown for filtering high frequency additive noise from a low spatial frequency image and for filtering a low frequency multiplicative noise from a high spatial frequency image.     

The Nonlinear Prefiltering and Difference of Estimates Approaches to Edge Detection: Applications of Stack Filters

The theory of stack filtering, which is a generalization of median filtering, is used in two different approaches to the detection of intensity edges in noisy images. The first approach is a generalization of median prefiltering: a stack filter or another median-type filter is used to smooth an image before a standard gradient estimator is applied. These prefiltering schemes retain the robustness of the median prefilter, but allow resolution of finer detail. The second approach, called the Difference of Estimates (DoE) approach, is a new formulation of a morphological scheme [Lee et al., IEEE Trans. Robotics Automat. RA-3, Apr. 1987, 142-156, Maragos and Ziff, IEEE Trans. Pattern Anal. Mach. Intell. 12(5), May 1990.] which has proven to be very sensitive to impulsive noise. In this approach, stack filters are applied to a noisy image to obtain local estimates of the dilated and eroded versions of the noise-free image. Thresholding the difference between these two estimates yields the edge map. We find, for example, that this approach yields results comparable to those obtained with the Canny operator for images with additive Gaussian noise, but works much better when the noise is impulsive. In both approaches, the stack filters employed are trained to be optimal on images and noise that are "typical" examples of the target image. The robustness of stack filters leads to good performance for the target image, even when the statistics of the noise and/or image vary from those used in training. This is verified with extensive simulations.     

A new fuzzy logic filter for image enhancement

This paper presents a new fuzzy-logic-control based filter with the ability to remove impulsive noise and smooth Gaussian noise, while, simultaneously, preserving edges and image details efficiently. To achieve these three image enhancement goals, we first develop filters that have excellent edge-preserving capability but do not perform well in smoothing Gaussian noise. Next, we modify the filters so that they perform all three image enhancement tasks. These filters are based on the idea that individual pixels should not be uniformly fired by each of the fuzzy rules. To demonstrate the capability of our filtering approach, it was tested on several different image enhancement problems. These experimental results demonstrate the speed, filtering quality, and image sharpening ability of the new filter.     

基于高斯马尔可夫随机场的MAP估计在高斯噪声滤波中的应用

提出了基于高斯马尔可夫随机场(GMRF)的最大后验概率(MAP)估计在图像高斯噪声滤波中的应用方法。根据高斯噪声的先验特点,建立基于高斯马尔可夫随机场的退化图像恢复模型,从而将图像高斯噪声滤波问题转化为求解最大后验概率问题。先验概率可以根据马尔可夫随机场(MRF)和吉布斯分布(GD)的等效性, 用GD的概率估计。为了求解最大后验概率,第一,通过期望最大化(EM）算法对GMRF模型进行参数估计。第二,用共轭梯度法将目标函数最小化。实验结果表明,与其他滤波器（如高斯滤波、维纳滤波等）相比,本文所阐述的方法在滤除高斯噪声、保持图像原有结构方面效果更好。     

Real-Time Color Image Processing Using Order Statistics Filters

The Vector Rank M-type K-Nearest Neighbour (VRMKNN) filter to remove impulsive noise from color images and video color sequences is presented. This filter utilizes multichannel image processing by using the vector approach and the Rank M-Type K-Nearest Neighbour (RMKNN) algorithm. Simulation results indicate that the proposed filter consistently outperforms other color image filters by balancing the tradeoff between noise suppression and detail preservation. The implementation of the filter was realized on the DSP TMS320C6711 to demonstrate that the proposed filter potentially could provide a real-time solution to quality video transmission.     

A Filter To Remove Gaussian Noise by Clustering the Gray Scale

An algorithm to suppress Gaussian noise is presented, based on clustering (grouping) gray levels. The histogram of a window sliding across the image is divided into clusters, and the algorithm outputs the mean level of the group containing the central pixel of the window. This filter restores well the majority of noisy pixels, leaving only few of them very deviated, that can be finally restored with a common filter for impulsive noise, such as a median filter. In this paper the clustering filter CF is described, analysed and compared with other similar filters.     

A parallel processing solution to the adaptive Kalman filtering problem with vector measurements

The adaptive Kalman filtering problem with vector measurements is considered. A computational algorithm is derived which gives estimates of the state of a linear dynamic system driven by additive white Gaussian noise with unknown covariance Q and observed by a linear function of the state contaminated by additive white Gaussian noise with unknown covariance R. The computational algorithm is inherently parallel in nature and it is noted that the algorithm should be implemented in a special purpose parallel processing digital computer made up of a number of filters similar to steady state Kalman filters each with a different gain. It is shown that the estimates of the state and the estimates of the unknown covariances Q and R can be made arbitrarily close to the optimal nonlinear Bayesian estimates by an appropriate choice for the number of parallel paths in the computer. When the filtering algorithm is implemented in a parallel processing computer the total processing time for state estimation in the unknown noise environment is only slightly increased over that required for a steady state Kalman filter. A numerical example with a five dimensional state and two dimensional measurement vector is presented.     

Iterative composite filtering for image restoration

An algorithm solution to the noisy image restoration problem under assumptions that the image is nonstationary and that the noise process is a superposition of white and impulsive noises is proposed. A composite model is used for the image in order to consider its nonstationarities, in the mean and the autocorrelation. Separating the gross information about the image from its textural information, the authors exploit the advantages of median, range, and Levinson filters in restoring the image. Median statistics are used to estimate the image's gross information and to filter the impulsive noise. Range statistics are used to segment the textural image into approximately locally stationary images to be filtered by Levenson filters. The proposed restoration algorithm adapts to the nonstationarity of the image, and, thus, it performs well. The algorithm is compared with others based on either median or linear filtering alone     

Efficient noise reduction in images using directional modified sigma filter

A noise reduction of images using a directional modified sigma filter is proposed. It is important that an image should include accurate values without noise for large-scale data processing of a cloud computing environment. A conventional sigma filter has been shown to be a good solution both in terms of filtering accuracy and computational complexity. However, the sigma filter does not preserve small edges well especially for the high level of additive noise. In this paper, we propose a new method using a modified sigma filter. In our proposed method, an input image is first decomposed into two components that have features of horizontal, vertical, and diagonal direction. Then two components are applied: high-pass filtering (HPF) and low-pass filtering (LPF). By applying the conventional sigma filter separately on each of them, an output image is reconstructed from the filtered components. Added noise is removed and our proposed method preserves the edges in the image. Comparative results from experiments show that the proposed algorithm achieves higher gains than the sigma filter and modified sigma filter, which are 2.6 dB PSNR on average and 0.5 dB PSNR, respectively. When relatively high levels of noise are added, the proposed algorithm shows better performance than the two conventional filters. The proposed method can be efficiently applied in digital cameras, digital TV, and smart phones.     

一种基于P2DHMT的小波域图像去噪算法

文章提出了一种基于小波域伪二维隐Markov树(P2DHMT)的图像的滤波新方法。首先建立了小波域的伪2DHMT模型,给出了基于EM、Baum-Welch等算法的模型参数估计方法;其次提出了一种基于最大后验概率准则的P2DHMT最优图像滤波算法;最后给出了图像去噪算法的实现过程。实验结果表明该方法可以在保存图像细节特征的情况下有效地抑制图像的噪声。     

基于小波收缩和正逆扩散结合的优质中值先验图像重建算法

针对最大后验(MAP)法对重建图像造成的过度平滑或出现阶梯状边缘伪影等问题,提出了一种基于混合模型的中值先验图像重建算法。首先在中值先验分布的MAP重建的基础上,在每次中值滤波之前引入结合小波收缩和正逆各向异性扩散的滤波器。另外,对于背景区域仍残留有少量噪声的情况下,可以在迭代间的最后,选择加入只针对图像较小梯度阈值区域进行非线性扩散平滑的优良滤波器,从而进一步优化图像。仿真结果表明,该算法在抑制噪声和保持边缘效果方面具有很好的表现,与其他经典传统算法相比,信噪比(SNR)可提高0.9dB~3.8dB。     

基于CSSD的非线性相关滤波在光照变化人脸识别中的应用

针对现有的滤波算法由于光照变化而影响人脸识别性能的问题,提出了特定类子空间依赖的非线性相关滤波算法。首先,利用非线性最佳映射图像相关滤波器与非线性最佳重建图像相关滤波器之间相位的特定类子空间运算实现算法;然后,通过最小化相关平面能量、同时最大化相关波峰进一步优化;最后,利用关联分类器完成人脸识别。在扩展Yale B和PIE人脸库上的实验结果表明,本文算法在加性高斯噪声条件下仍然对光照变化不敏感,相比其他几种较好的滤波算法,本文算法取得了更高的识别率,并提高了算法执行效率。     

A new adaptive center weighted median filter for suppressing impulsive noise in images

In this work, a new adaptive center weighted median (ACWM) filter is proposed for improving the performance of median-based filters, preserving image details while effectively suppressing impulsive noise. The proposed filter is an adaptive CWM filter with an adjustable central weight obtained by partitioning the observation vector space. To obtain the optimal weight for each block, the efficient scalar quantization (SQ) method is used to partition the observation vector space. The center weight within each block is obtained by using a learning approach based on the least mean square (LMS) algorithm. The noise filtering procedure is progressively applied through several iterations so that the mean square error of the output can be minimized. Experimental results have demonstrated that the proposed filter outperforms many well-accepted median-based filters in terms of both noise suppression and detail preservation. The proposed new filter also provides excellent robustness at various percentages of impulsive noise.     

Improved bilateral filter for suppressing mixed noise in color images

It is a challenging problem to suppress mixed noise in color images. The traditional bilateral filter can excellently reduce additive noise without destroying image edges and details, but it fails to remove impulsive noise. This paper presents an improved bilateral filtering method, which can simultaneously suppress both impulsive and additive noise. The proposed solution first introduces a new weighting function to the bilateral filtering mechanism, which is experimentally more effective than the traditional Gaussian kernel. Then, either the current pixel or the vector median, instead of always the current pixel itself, is chosen as the base to take part in the bilateral filtering action, which is determined by whether the current pixel is a possible impulse or not. The experimental results show that the proposed solution can simultaneously remove impulsive and additive noise while preserving edge structures, and outperforms other vector filtering methods in terms of both objective evaluation and subjectively visual quality.     

基于四元数旋转的开关型矢量中值滤波器

矢量中值滤波器是一种经典和高效的矢量滤波器,主要用于消除彩色图像中的冲击噪声。然而VMF没有区分细线条和噪声的能力,它往往把细线条当成噪声而过滤掉。本文利用四元数旋转理论,模仿Laplacian算子,提出一种用于检测彩色图像中的冲击噪声的算法,并结合传统的VMF构造出一个新颖的开关型矢量中值滤波器。实验结果表明,新的滤波器不仅能有效地保护细线条和边界等细节信息,而且其滤波性能也明显胜过传统的VMF和一些经典的及最近开发的矢量滤波器。     

一种改进的脉冲噪声滤波方法

由于数字图像在生成与传输过程中容易受到脉冲噪声的污染,往往造成后续处理难以为继。为了改善图像质量,需要对图像进行去噪处理。针对传统中值滤波及其它非线性滤波方法在去除图像脉冲噪声时存在的不足,本文提出了一种改进的去噪方法:在滤波之前进行一次脉冲噪声检测,确定受到噪声污染的像素点,并进行记录标识;然后根据检测结果进行改进的中值滤波:只对判断为噪声点的像素进行处理,不仅考虑了标准中值,也分情况利用了中值的前一个值和中值的后一个值的信息。实验表明,改进方法不仅在滤除脉冲噪声方面相比其他非线性滤波有很大改进,而且它可以更好地保护图像的细节特性,对图像的后续处理有很好的价值。     

基于模糊推理的多窗口中央值滤波器设计方法

提出了一种基于模糊推理用于去除图像椒盐噪声的中央值滤波器的新型设计方法,在图像复原处理中,理想的期望是对图像被劣化的部分处理,没有被劣化的部分不作处理,但实际图像处理中处理点是否为噪声点具有模糊性.利用模糊推理对处理点像素多大程度上属于劣质像素进行推定,并且多个模糊滤波器联合使用,处理结果证明对广范围噪声发生率的各种被椒盐噪声劣化的图像复原处理都适用.     

一种彩色图像滤波的改进矢量中值滤波算法

通过分析彩色图像滤波的一般方法，提出了一种基于矢量中值滤波的改进算法。该算法融合了线性均值滤波和非线性矢量中值滤波两种方法，大大降低了运算复杂度，同时对脉冲噪声和高斯噪声有好的抑制作用，能有效地保护图像的边缘信息，滤波后不会出现新的颜色。另外本算法的运算复杂度随着滑动窗的增大而缓慢增加，使其能够达到滤波效果和运算复杂度的有效平衡。