A class of adaptive directional image smoothing filters

The gray level distribution around a pixel of an image usually tends to be more coherent in some directions compared to other directions. The idea of adaptive directional filtering is to estimate the direction of higher coherence around each pixel location and then to employ a window which approximates a line segment in that direction. Hence, the details of the image may be preserved while maintaining a satisfactory level of noise suppression performance. In this paper we describe a class of adaptive directional image smoothing filters based on generalized Gaussian distributions. We propose a measure of spread for the pixel values based on the maximum likelihood estimate of a scale parameter involved in the generalized Gaussian distribution. Several experimental results indicate a significant improvement compared to some standard filters.     

一种异性扩散的自适应平滑滤波算法

固定尺度的高斯滤波器对图像中的所有像素使用相同的加权系数,在平滑噪声的同时也模糊了边缘信息,影响了后期的边缘定位和提取,为此提出了一种异性扩散的自适应仿高斯滤波器,通过计算当前像素与周围像素的灰度梯度,预判该像素类型（噪声点或是边缘点）,并根据像素点类型和梯度生成自适应的、各向异性的加权系数,通过多次平滑迭代实现权重的异性扩散。实验结果表明,该滤波算法具有平滑噪声的同时保留并锐化边缘效果。     

An effective edge-preserving smoothing method for image manipulation

This paper presents a novel and effective edge-preserving image smoothing method for edge-aware image manipulation. The method formulates the smoothing as a problem of minimizing a convex object function with a constraint and an efficient solution to the optimization problem is presented. Specifically, the method provides an unified framework to regularize the edge and texture pixels in the optimization so that geometric edges representing image structures can be well retained and fine edges of texture regions are removed or suppressed. Both qualitative and quantitative experimental results on natural images and computer-generated structured images have shown the efficacy of the proposed method. In addition, the proposed method can improve the performance of many image processing and manipulation tasks including edge extraction and simplification, non-photorealistic rendering, detail and contrast exaggeration, HDR tone mapping, block-based discrete cosine transform (BDCT) artifact removal and content-aware image resizing, as demonstrated through the experiments.     

A novel adaptive SVR based filter ASBF for image restoration

In this paper, a novel adaptive filter ASBF based on support vector regression (SVR) is proposed to preserve more image details and efficiently suppress impulse noise simultaneously. The main idea of the novel filter ASBF here is to employ a SVR based impulse detector to judge whether an input pixel is contaminated or not by impulse noise. If this case happens, a median filter is employed to remove the corresponding impulse noise. This judgment procedure is executed by regressing the filter window of an input pixel using SVR and then judging the input pixel by its regression distance. Huber loss function is used in SVR regression, due to its excellent robustness capability. The distinctive advantage of the filter ASBF over the latest Support Vector Classifier (SVC) based filter is that no training for the original noise-free image is required in our approach, which is well in accordance with our visual judgment way. Experimental results for benchmark images demonstrate that our filter ASBF here outperforms the extensively-used median-based filters and the SVC based filter.     

An adaptive high-gain observer for nonlinear systems

In this paper the authors provide a solution to the noise sensitivity of high-gain observers. The resulting nonlinear observer possesses simultaneously (1) extended Kalman filter’s good noise filtering properties, and (2) the reactivity of the high-gain extended Kalman filter with respect to large perturbations.The authors introduce innovation as the quantity that drives the gain adaptation. They prove a general convergence result, propose guidelines to practical implementation and show simulation results for an example.     

一种改进的非线性图像滤波器

基于均值滤波器提出了一种新的非线性滤波器,该滤波器可以同时处理高斯噪声和椒盐噪声.提高了非线性滤波器对复杂噪声环境的适应能力.     

An adaptive image interpolation algorithm for image/video processing

Image interpolation is one of the key technologies in image/video processing. In this study, a new adaptive image interpolation algorithm is proposed. The objective of the proposed approach is to recover up-sampled image frames from the corresponding decimated (low-resolution) image frames. In the proposed approach, within each iteration, two proposed nonlinear filters are utilized to iteratively generate high-frequency components lost within the decimation procedure. Finally, a post-processing procedure is adopted to reduce the blocking artifacts within the interpolated images. Based on the experimental results obtained in this study, in terms of the average PSNR_p (peak signal-to-noise ratio) in dB and subjective measure of the quality of the interpolated images, the interpolation results by the proposed approach are better than that by three existing interpolation approaches for comparison.     

图像降噪的自适应高斯平滑滤波器

作为去除图像中噪声的图像增强技术,常用的图像平滑方法在提高局部信噪比的同时,也使图像产生模糊。为克服上述缺点,引入了自适应高斯滤波器,它结合了高斯滤波器和梯度倒数加权滤波器的特点,同时考虑了图像局部的空间距离和像素距离,以确定参与局部平滑的像素及其权值。该滤波器算法牺牲了简单平滑滤波器的计算性能,但很好地保留了图像的局部特点,特别是边缘和细节。实验比较了该方法与其他常用滤波器的性能,结果证实了该方法的有效性。     

Graph spectral image smoothing using the heat kernel

A new method for smoothing both gray-scale and color images is presented that relies on the heat diffusion equation on a graph. We represent the image pixel lattice using a weighted undirected graph. The edge weights of the graph are determined by the Gaussian weighted distances between local neighboring windows. We then compute the associated Laplacian matrix (the degree matrix minus the adjacency matrix). Anisotropic diffusion across this weighted graph-structure with time is captured by the heat equation, and the solution, i.e. the heat kernel, is found by exponentiating the Laplacian eigensystem with time. Image smoothing is accomplished by convolving the heat kernel with the image, and its numerical implementation is realized by using the Krylov subspace technique. The method has the effect of smoothing within regions, but does not blur region boundaries. We also demonstrate the relationship between our method, standard diffusion-based PDEs, Fourier domain signal processing and spectral clustering. Experiments and comparisons on standard images illustrate the effectiveness of the method.     

保持边缘的高斯平滑滤波算法研究

提出了一种新的彩色图像高斯滤波算法EPGF,该方法从人眼视觉感知中的彩色细节失明及同时对比特性出发,根据像素间的色差和位置关系共同确定高斯权值,使算法对区域内部进行平滑的同时,保持区域间的边缘不变。给出了各种滤波结果的主观和客观的对比,实验结果表明了算法的有效性。     

Easy recipes for cooperative smoothing

In this paper we suggest how several competing signal smoothers, differing in design parameters, or even in design principles, can be combined together to yield a better and more reliable smoothing algorithm. The proposed heuristic, but statistically well motivated, fusion mechanism allows one to combine practically all kinds of smoothers, from simple local averaging or order statistic filters, to parametric smoothers designed for different hypothetical signal and/or noise models. It also allows one to account for the distribution of measurement noise, and in particular to cope with heavy-tailed disturbances, such as Laplacian noise, and light-tailed disturbances, such as uniform noise.     

Edge-preserving smoothing using a similarity measure in adaptive geodesic neighbourhoods

This paper introduces a novel image-dependent filtering approach derived from concepts known in mathematical morphology and aiming at edge-preserving smoothing of natural images. Like other adaptive methods, it assumes that the neighbourhood of a pixel contains the essential information required for the estimation of local features in the image. The proposed strategy essentially consists in a weighted averaging combining both spatial and tonal information. For that purpose, a twofold similarity measure is calculated from local geodesic time functions. This way, no prior operator definition is required since a weighting neighbourhood and a weighting kernel are determined automatically from the unfiltered input data for each pixel location. By designing relevant geodesic masks, two adaptive filtering algorithms are derived that are particularly efficient at smoothing heterogeneous areas while preserving relevant structures in greyscale and multichannel images.     

一种改进的自适应中值滤波方法

提出了一种改进的自适应中值滤波算法,以有效地去除图像中的脉冲噪声,并保留图像细节。在进行噪声点检测时,引入了最小集合距离测度,有效地避免了将高频细节信号误判为噪声。采用最小无污染点集合的中值恢复噪声点,消除了其邻域噪声点的影响。通过与RAMF、NASMF等方法的比较实验表明,新算法噪声检测的正确率高、降噪与保留细节效果好, 尤其对含噪声密度高的图像的处理效果优势更为明显。     

An adaptive image Euclidean distance

The image Euclidean distance (IMED) considers the spatial relationship between the pixels of different images and can easily be embedded in existing image recognition algorithms that are based on Euclidean distance. IMED uses the prior knowledge that pixels located near one another have little variance in gray scale values, and defines a metric matrix according to the spatial distance between pixels. In this paper, we propose an adaptive image Euclidean distance (AIMED), which considers not only the prior spatial knowledge, but also the prior gray level knowledge from images. The most important advantage of the proposed AIMED over IMED is that AIMED makes the metric matrix adaptive to the content of the concerned images. Two ways of using gray level information are proposed. One is based on gray level distances, and the other is based on cosine dissimilarity of gray levels. Experiments on two facial databases and a handwritten digital database show that AIMED achieves the highest classification accuracy when it is embedded in nearest neighbor classifiers, principal component analysis, and support vector machines.     

基于自适应Mean Shift算法的彩色图像滤波

利用Mean Shift算法对图像进行滤波时,带宽和采样点权重的选择至关重要。为有效保留彩色图像中边缘等细节信息,提出了一种基于自适应Mean Shift算法的彩色图像滤波算法。该算法首先根据图像颜色信息的灰色关联度来定义自适应空域带宽;然后利用颜色粗糙度计算自适应值域带宽;再根据彩色差别获得各采样点的权重;最后用改进后的自适应Mean Shift算法对图像进行滤波。实验结果表明,与基于Mean Shift算法的滤波算法和常见滤波算法相比,该算法的滤波效果不仅更符合人眼的视觉感知,且能更好地保留边缘等细节信息。     

图像平滑的2维小波插值方法

在应用扩散方程进行图像平滑时,常规的方法是对扩散方程差分化构造差分方程,利用初边值条件求解。这种方法误差传播快,精度不高。因此,构造了2维小波插值函数,利用它来求解扩散方程,并分析得到用小波插值函数求解Alvarez模型的方法。由于小波函数具有良好的局部性,求解扩散方程比用差分方法求解具有精度高,误差传播速度慢,对时间步长不敏感等优点。在数值实验中,给出了本文方法的有效性及相对于差分方法求解的优点。     

各向异性扩散的彩色图像复合滤波方法

将扩散偏微分方程PDE’s推广到彩色图像滤波。分析了基于散度算子和基于迹算子PDE的优缺点,提出了一个新的扩散PDE方法。该方法在各向异性扩散基础上增加了一个震动滤波算子,并对特征根和扩散张量进行了重构,不仅对图像有很好的滤波效果,而且对边缘有增强保护的作用。实验结果表明,提出的方法达到了既有效去除噪声又保留更多细节的目的。     

Predicting the Brazilian stock market through neural networks and adaptive exponential smoothing methods

The study of financial markets has been addressed in many works during the last years. Different methods have been used in order to capture the non-linear behavior which is characteristic of these complex systems. The development of profitable strategies has been associated with the predictive character of the market movement, and special attention has been devoted to forecast the trends of financial markets. This work performs a predictive study of the principal index of the Brazilian stock market through artificial neural networks and the adaptive exponential smoothing method, respectively. The objective is to compare the forecasting performance of both methods on this market index, and in particular, to evaluate the accuracy of both methods to predict the sign of the market returns. Also the influence on the results of some parameters associated to both methods is studied. Our results show that both methods produce similar results regarding the prediction of the index returns. On the contrary, the neural networks outperform the adaptive exponential smoothing method in the forecasting of the market movement, with relative hit rates similar to the ones found in other developed markets.     

基于纹理分析的保细节平滑滤波器

平滑去噪是图象处理中一个重要课题,但是以往在处理平滑去噪问题上一直存在平滑和保细节的矛盾。为解决此问题,提出了一种基于纹理分析和保细节平滑滤波器,该滤波器采用了多尺度多方向的模板,并利用纹理分析等手段,同时根据图象各部分特性,通过自适应地选择模板来进行平滑滤波,该算法兼顾了降噪和保细节两方面要求。实验结果证明,该算法实现简单,计算速度快,且效果优于其他几种常用的保边界平滑算法。     

自适应平滑自商图与多层分块的人眼定位算法

针对光照对眼睛定位产生的影响,提出了一种基于自适应平滑自商图像与多层分块的人眼定位算法。首先进行光照评估,利用自适应平滑自商图像（ASSQI）对光照条件差的进行处理,消除人脸图像的光照影响;然后运用眼睛区域梯度复杂度大的特征并结合积分图分块处理获得眼睛粗定位;最后对粗定位区域继续分块并利用质心得到准确的瞳孔中心。在JAFFE数据库、Yale B数据库和BioID数据库上测试表明该文算法有较高的定位正确率,对光照的鲁棒性较强。