SVD-based modeling for image texture classification using wavelet transformation.

This paper introduces a new model for image texture classification based on wavelet transformation and singular value decomposition. The probability density function of the singular values of wavelet transformation coefficients of image textures is modeled as an exponential function. The model parameter of the exponential function is estimated using maximum likelihood estimation technique. Truncation of lower singular values is employed to classify textures in the presence of noise. Kullback-Leibler distance (KLD) between estimated model parameters of image textures is used as a similarity metric to perform the classification using minimum distance classifier. The exponential function permits us to have closed-form expressions for the estimate of the model parameter and computation of the KLD. These closed-form expressions reduce the computational complexity of the proposed approach. Experimental results are presented to demonstrate the effectiveness of this approach on the entire 111 textures from Brodatz database. The experimental results demonstrate that the proposed approach improves recognition rates using a lower number of parameters on large databases. The proposed approach achieves higher recognition rates compared to the traditional sub-band energy-based approach, the hybrid IMM/SVM approach, and the GGD-based approach.     

Speckle removal from SAR images in the undecimated wavelet domain

Speckle reduction is approached as a minimum mean-square error (MMSE) filtering performed in the undecimated wavelet domain by means of an adaptive rescaling of the detail coefficients, whose amplitude is divided by the variance ratio of the noisy coefficient to the noise-free one. All the above quantities are analytically calculated from the speckled image, the variance and autocorrelation of the fading variable, and the wavelet filters only, without resorting to any model to describe the underlying backscatter. On the test image Lena corrupted by synthetic speckle, the proposed method outperforms Kuan's local linear MMSE filtering by almost 3-dB signal-to-noise ratio. When true synthetic aperture radar (SAR) images are concerned, empirical criteria based on distributions of multiscale local coefficient of variation, calculated in the undecimated wavelet domain, are introduced to mitigate the rescaling of coefficients in highly heterogeneous areas where the speckle does not obey a fully developed model, to avoid blurring strong textures and point targets. Experiments carried out on widespread test SAR images and on a speckled mosaic image, comprising synthetic shapes, textures, and details from optical images, demonstrate that the visual quality of the results is excellent in terms of both background smoothing and preservation of edge sharpness, textures, and point targets. The absence of decimation in the wavelet decomposition avoids typical impairments often introduced by critically subsampled wavelet-based denoising.     

深度成像理论与实现

传统成像获取信息不足,成像质量有一定局限性。为此,提出了一种深度成像模型。模型包含深度矩阵、分解函数、散焦算子、自适应正则项等部分。深度矩阵的获取有双目立体视觉、结构光或飞行时间法等实现方法;分解函数用于将图像按深度值的不同分割为若干子图像;散焦算子可以通过深度散焦法来计算;自适应正则项的引入能减少图像的阶梯效应,增强图像的光滑性。通过局部标准差和局部平均梯度这两个评价指标检验深度成像模型的效果。实验结果表明,深度成像模型效果显著。     

SAR图像去噪的分数阶多尺度变分PDE模型及自适应算法

在合成孔径雷达(SAR)图像相干斑噪声抑制中,保持图像的边缘和纹理是非常重要的。该文首先利用分数阶导数和负指数Sobolev空间对图像进行建模,建立了分数阶多尺度变分偏微分方程(PDE)模型,然后给出了模型参数自适应选择方法,并在此基础上提出了区域、尺度自适应的去噪算法。数值实验表明,新方法能在去除噪声,抑制图像的 阶梯效应,保持图像的边缘、纹理细节几个方面取得较好的效果。     

基于FRGF和改进PCNN的红外可见光图像融合

为解决图像融合中边缘细节保留不理想的问题,本文提出了一种快速滚动引导滤波器和改进脉冲耦合神经网络相结合的红外可见光图像融合方法。提出的快速滚动引导滤波器可以较好地在保留边缘、细节纹理信息的同时有效提高运行效率。首先,利用快速滚动引导滤波和高斯滤波对源图像进行多尺度分解;其次,为了使基础层图像更好地突出轮廓信息,采用相似性匹配的融合规则对图像进行融合;然后,细节层采用改进参数自适应脉冲耦合神经网络规则进行融合;最后,经过多尺度重构得到融合结果图。实验结果表明,与其它5种融合方法相比,该算法不仅在视觉效果上得到了提升,而且能够充分保存图像的边缘和纹理等信息,极大地提高了运行效率。另外,该方法在客观评价指标上均优于对比方法。     

On texture and image interpolation using Markov models

Markov-type models characterize the correlation among neighboring pixels in an image in many image processing applications. Specifically, a wide-sense Markov model, which is defined in terms of minimum linear mean-square error estimates, is applicable to image restoration, image compression, and texture classification and segmentation. In this work, we address first-order (auto-regressive) wide-sense Markov images with a separable autocorrelation function. We explore the effect of sampling in such images on their statistical features, such as histogram and the autocorrelation function. We show that the first-order wide-sense Markov property is preserved, and use this result to prove that, under mild conditions, the histogram of images that obey this model is invariant under sampling. Furthermore, we develop relations between the statistics of the image and its sampled version, in terms of moments and generating model noise characteristics. Motivated by these results, we propose a new method for texture interpolation, based on an orthogonal decomposition model for textures. In addition, we develop a novel fidelity criterion for texture reconstruction, which is based on the decomposition of an image texture into its deterministic and stochastic components. Experiments with natural texture images, as well as a subjective forced-choice test, demonstrate the advantages of the proposed interpolation method over presently available interpolation methods, both in terms of visual appearance and in terms of our novel fidelity criterion.     

参数自适应的耦合卡通-纹理分解及边缘检测方法

在耦合卡通-纹理分解及边缘检测模型的基础上,提出了根据原图像在不同位置的图像特征自适应选取参数的算法,即根据对原图像预分析的结果,分别在边缘、纹理、分片光滑区域选取不同的参数值,从而达到充分提取纹理、保证卡通图像光滑性同时保护边缘的目的。数值实验结果表明,本文方法提高了图像分解的质量及边缘检测的准确性,减少了边缘提取的奇异点。     

Blind adaptive sampling of images

Adaptive sampling schemes choose different sampling masks for different images. Blind adaptive sampling schemes use the measurements that they obtain (without any additional or direct knowledge about the image) to wisely choose the next sample mask. In this paper, we present and discuss two blind adaptive sampling schemes. The first is a general scheme not restricted to a specific class of sampling functions. It is based on an underlying statistical model for the image, which is updated according to the available measurements. A second less general but more practical method uses the wavelet decomposition of an image. It estimates the magnitude of the unsampled wavelet coefficients and samples those with larger estimated magnitude first. Experimental results show the benefits of the proposed blind sampling schemes.     

结合Contourlet变换的Bayes自适应图像去噪算法

提出了结合Contourlet变换的Bayes自适应图像去噪算法。充分利用Contourlet变换的局部性、多分辨率、各向异性等优点。通过Contourlet变换得到图像不同尺度不同方向上Contourlet系数矩阵,实现了建立在对图像多尺度几何分析基础上Bayes估计自适应图像去噪算法。实验表明,新算法能够获得良好的视觉效果并且有效地提高了去噪图像的PSNR值,同时有效的避免了“过扼杀”系数现象,更好地保留了图像的纹理和细节。     

Speckle filtering of SAR images based on adaptive windowing

Speckle noise usually occurs in synthetic aperture radar (SAR) images owing to coherent processing of SAR data. The most well-known image domain speckle filters are the adaptive filters using local statistics such as the mean and standard deviation. The local statistics filters adapt the filter coefficients based on data within a fixed running window. In these schemes, depending on the window size, there exists trade-off between the extent of speckle noise suppression and the capability of preserving fine details. The authors propose a new adaptive windowing algorithm for speckle noise suppression which solves the problem of window size associated with the local statistics adaptive filters. In the algorithm, the window size is automatically adjusted depending on regional characteristics to suppress speckle noise as much as possible while preserving fine details. Speckle noise suppression gets stronger in homogeneous regions as the window size increases succeedingly. In fine detail regions, by reducing the window size successively, edges and textures are preserved. The fixed-window filtering schemes and the proposed one are applied to both a simulated SAR image and an ERS-1 SAR image to demonstrate the excellent performance of the proposed adaptive windowing algorithm for speckle noise     

Texture analysis-based multi-focus image fusion using a modified Pulse-Coupled Neural Network (PCNN)

Multi-focus image fusion is an effective method of information fusion that can take a series of source images and obtain a fused image where everything is in focus. In this paper, a multi-focus image fusion method based on image texture that adopts a modified Pulse-Coupled Neural Network (PCNN) approach is proposed. First, the texture of an image is obtained by means of image cartoon and texture decomposition. An ignition image is then acquired by inputting the image textures into a modified PCNN. Ignition images are compared to each other to obtain an initial decision map. A small object detection and bilateral filter is then applied to the initial decision map to reduce noise and enable smoother processing. Finally, the source images and decision map are used to produce the fused image. Experimental results demonstrate that the proposed method effectively preserves the source images information while delivering good image fusion performance.     

自适应阈值及加权局部二值模式的人脸识别

针对局部二值模式(LBP)和中心对称局部二值模式(CS-LBP)方法描述图像纹理特征时,阈值不能自动选取并且图像中不同子块的贡献也没有进行区分的问题,该文提出一种自适应阈值及加权的局部二值模式方法。首先,将图像进行分块,采用设定的自适应阈值提取每个子块的LBP或CS-LBP纹理直方图;然后,将各子图像的信息熵作为直方图的加权依据,对每个子块对应的直方图进行自适应加权,并将所有子块的直方图连接成最终的纹理特征;最后,通过快速计算图像均值加快了算法的计算速度。在人脸数据库上进行的实验证明,利用该文提出的方法提取纹理特征,并结合最近邻分类法可以得到较高的正确识别率。     

基于波前相位单纯形样条函数建模的空间目标波前解卷积方法

针对空间目标地基自适应光学望远镜成像过程中同时记录目标图像及波前传感器数据的情形,提出了一种利用波前测量数据的空间目标自适应光学图像复原方法。该方法将大气降质波前表示为望远镜孔径内的二元单纯形样条函数,而非传统的Zernike模式的线性组合,基于该区域表示,为哈特曼-夏克波前传感器建立了平均斜率测量模型,进而非适定的波前重构问题转化为良态等式约束最小二乘问题,最终的目标图像即可通过非盲解卷积方法获得。仿真实验验证了提出的方法在不同的湍流强度下均能表现出良好的复原效果,对测量噪声亦不敏感。     

A restoration algorithm for images contaminated by mixed Gaussian plus random-valued impulse noise

In this paper, we study the problem of restoring the image corrupted by additive Gaussian noise plus random-valued impulse noise. A novel noise classifier is firstly created to identify different noise in the corrupted image. Then, we use the remaining effective information to train an adaptive overcomplete dictionary for sparse representation of image patches with the help of masked K-SVD algorithm. Because of the adaptive nature of the learned dictionary, it can represent the image patches in concern more efficiently. Then, we minimize a variational model containing an optional data-fidelity term and a smooth regularization term respecting sparse representation of every image patch to get the final restored image. Extensive experimental results prove that our method cannot only remove noise from the corrupted image well, but also preserve more details and textures. It surpasses some state-of-the-art methods.     

New prediction schemes for lossless coding of fullband and subband images

This paper presents a study of lossless image compression of fullband and subband images using predictive coding. The performance of a number of different fixed and adaptive predictors are evaluated to establish the relative performance of different predictors at various resolutions and to give an indication of the achievable image resolution for given bit rates. In particular, the median adaptive predictor is compared with two new classes of predictors proposed in this paper. One is based on the weighted median filter, while the other uses context modelling to select the optimum from a set of predictors. A graphical tool is also proposed to analyse the prediction methods. Simulations of the different predictors for a variety of real world and medical images, evaluated both numerically and graphically, show the superiority of median based prediction over this proposed implementation of context model based prediction, for all resolutions. The effects of different subband decomposition techniques are also explored.     

基于OSV分解的图像多尺度特征表示

 本文给出一种新的图像多尺度表示算法.首先,应用OSV模型得到图像的单尺度分解;其次,针对上一步的信息亏损,引入不同的单调尺度参数,迭代OSV变分模型,从而为图像的不同特征提供一种非线性的分级自适应表达式.同时,本文也给出有关新算法的离散格式.数值实验表明,与已有的Nezzar算法相比,新算法的多尺度分解效果更佳.     

Shannon小波包分解自适应Gabor滤波器设计及其在纹理分割中的应用

本文首先研究了Ｇａｂｏｒ滤波器进行确定性纹理分割时纹理结构与Ｇａｂｏｒ函数的关系，提出了利用在频域中具有紧支集的Ｓｈａｎｎｏｎ小波包分解检测纹理的主频；最后，提出了基于小波包分解自适应Ｇａｂｏｒ函数与纹理图像卷积，就可以在纹理的连续处产生良好的阶跃边缘，并通过实验验证了该算法的有效性及其对噪声的鲁棒性。     

变分框架下的多尺度图像恢复与重建

变分图像分解,通过极小化能量泛函将图像分解为不同的特征分量,可以被应用到图像的恢复和重建.提出了变分框架下的多尺度图像恢复和重建的思想.基于这种思想,首先提出了一个单参数的（BV,G,E）三元变分分解模型,并且理论分析了参数与不同特征分量的尺度的关系.然后将此模型的参数选为一个二进制序列,得到多尺度的（BV,G,E）变分分解.该多尺度变分分解可以将图像分解为一序列图像结构、纹理和噪声.证明了此多尺度分解的收敛性并且基于对偶理论和交替迭代算法给出了其数值求解方法.最后将提出的多尺度的（BV,G,E）变分分解应用到图像恢复和重建,实验结果证实了理论分析的正确性,显示了将此模型进行图像多尺度恢复和重建的有效性,和与一些其他分解模型相比较的优越性.     

改进的PCNN模型在多光谱与全色图像融合中的应用研究

介绍了PCNN模型原理,提出了基于双通道自适应的PCNN多光谱与全色图像融合算法。该算法首先将RGB空间的多光谱图像转换为HSV彩色空间,然后将HSV彩色空间中的非彩色通道(V通道)的灰度像素值和全色图像的像素灰度值分别作为PCNN-1及PCNN-2的神经元输入,利用方向性信息作为自适应链接强度系数,对非彩色通道图像和全色图像进行自适应分解,再将点火时间序列送入判决因子得到新的非彩色通道图像,最后将原多光谱图像的H通道分量、S通道分量及新的V通道分量经HSV空间逆变换获得最终的融合图像。实验结果表明,该算法不仅解决了链接强度系数自动设置的问题,而且充分考虑到图像边缘和方向特征的影响,无论在主观视觉效果,还是客观评价标准上均优于IHS、PCA、小波融合等其他图像融合算法,同时降低了计算复杂度。     

一种基于迭代映射和图像内容的自适应水印算法

提出了一种新颖的基于迭代映射和图像内容的自适应水印算法。该算法将图像分割为互不重叠的图像块,利用分形维数分析图像块内容,选出反映图像自身特征的边缘和纹理区域。再根据图像块内部以及相邻图像块之间的方差特性将所选的特征区域细分为边缘、弱纹理和强纹理。对选取的特征块进行离散余弦变换并将经过迭代映射置乱预处理的灰度数字水印以不同的强度自适应地嵌入到中频系数中。实验结果表明该算法透明性好,安全性高,对剪切、低通和高通滤波、添加噪声和纹理以及JPEG压缩等具有较强的顽健性。