Segmentation of Gabor-filtered textures using deterministicrelaxation

A supervised texture segmentation scheme is proposed in this article. The texture features are extracted by filtering the given image using a filter bank consisting of a number of Gabor filters with different frequencies, resolutions, and orientations. The segmentation model consists of feature formation, partition, and competition processes. In the feature formation process, the texture features from the Gabor filter bank are modeled as a Gaussian distribution. The image partition is represented as a noncausal Markov random field (MRF) by means of the partition process. The competition process constrains the overall system to have a single label for each pixel. Using these three random processes, the a posteriori probability of each pixel label is expressed as a Gibbs distribution. The corresponding Gibbs energy function is implemented as a set of constraints on each pixel by using a neural network model based on Hopfield network. A deterministic relaxation strategy is used to evolve the minimum energy state of the network, corresponding to a maximum a posteriori (MAP) probability. This results in an optimal segmentation of the textured image. The performance of the scheme is demonstrated on a variety of images including images from remote sensing.     

Image denoising via local and nonlocal circulant similarity

A patch based image denoising method is developed in this paper by introducing a new type of image self-similarity. This self-similarity is obtained by cyclic shift, which is called “circulant similarity”. Given a corrupted image patch, it can be estimated by incorporating circulant similarity into a weighted averaging filter. By choosing an appropriate kernel as weight function, the patch filter is implemented by circular convolution, and can be efficiently solved using fast Fourier transform. In addition, the circulant similarity can be enhanced by using nonlocal modeling. We stack the similar image patches into 3D groups, and propose a denoising scheme based on group estimation across the patches. Numerical experiments demonstrate that the proposed method with local circulant similarity outperforms much its local filtering based counterparts, and the proposed method with nonlocal circulant similarity shows very competitive performance with state-of-the-art denoising method, especially on images corrupted by strong noise.     

An image denoising algorithm for mixed noise combining nonlocal means filter and sparse representation technique

Nonlocal means (NLM) filtering or sparse representation based denoising method has obtained a remarkable denoising performance. In order to integrate the advantages of two methods into a unified framework, we propose an image denoising algorithm through skillfully combining NLM and sparse representation technique to remove Gaussian noise mixed with random-valued impulse noise. In the non-Gaussian circumstance, we propose a customized blockwise NLM (CBNLM) filter to generate an initial denoised image. Based on it, we classify the different noisy pixels according to the three-sigma rule. Besides, an overcomplete dictionary is trained on the initial denoised image. Then, a complementary sparse coding technique is used to find the sparse vector for each input noisy patch over the overcomplete dictionary. Through solving a more reasonable variational denoising model, we can reconstruct the clean image. Experimental results verify that our proposed algorithm can obtain the best denoising performance, compared with some typical methods.     

Robust NL-Means Filter With Optimal Pixel-Wise Smoothing Parameter for Statistical Image Denoising

Most denoising methods require that some smoothing parameters be set manually to optimize their performance. Among these methods, a new filter based on nonlocal weighting (NL-means filter) has been shown to have a very attractive denoising capacity. In this paper, we propose fixing the smoothing parameter of this filter automatically. The smoothing parameter corresponds to the bandwidth h of a local constant regression. We use the C_p statistic embedded in Newton's method to optimize h in a point-wise fashion. This statistic also has the advantage of being a reliable measure of the quality of the denoising process for each pixel. In addition, we introduce a robust regression in the NL-means filter designed to greatly reduce the blur yielded by the weighting. Finally, we show how the automatic denoising model can be extended to images degraded by multiplicative noise. Experiments conducted on images with additive and multiplicative noise demonstrate a high denoising power with a degree of detail preservation...     

Gaussian principle components for nonlocal means image denoising

NonLocal Means (NLM), taking fully advantage of image redundancy, has been proved to be very effective in noise removal. However, high computational load limits its wide application. Based on Principle Component Analysis (PCA), Principle Neighborhood Dictionary (PND) was proposed to reduce the computational load of NLM. Nevertheless, as the principle components in PND method are computed directly from noisy image neighborhoods, they are prone to be inaccurate due to the presence of noise. In this paper, an improved scheme for image denoising is proposed. This scheme is based on PND and uses preprocessing via Gaussian filter to eliminate the influence of noise. PCA is then used to project those filtered image neighborhood vectors onto a lower-dimensional space. With the preprocessing process, the principle components computed are more accurate resulting in an improved denoising performance. A comparison with some NLM based and state-of-art denoising methods shows that the proposed method performs well in terms of Peak Signal to Noise Ratio (PSNR) as well as image visual fidelity. The experimental results demonstrate that our method outperforms existing methods both subjectively and objectively.     

Unsupervised texture classification using vector quantization anddeterministic relaxation neural network

This paper describes the use of a neural network architecture for classifying textured images in an unsupervised manner using image-specific constraints. The texture features are extracted by using two-dimensional (2-D) Gabor filters arranged as a set of wavelet bases. The classification model comprises feature quantization, partition, and competition processes. The feature quantization process uses a vector quantizer to quantize the features into codevectors, where the probability of grouping the vectors is modeled as Gibbs distribution. A set of label constraints for each pixel in the image are provided by the partition and competition processes. An energy function corresponding to the a posteriori probability is derived from these processes, and a neural network is used to represent this energy function. The state of the network and the codevectors of the vector quantizer are iteratively adjusted using a deterministic relaxation procedure until a stable state is reached. The final equilibrium state of the vector quantizer gives a classification of the textured image. A cluster validity measure based on modified Hubert index is used to determine the optimal number of texture classes in the image.     

基于非局部均值的彩色图像去噪

传统的彩色图像去噪算法通常是分层处理的,而忽略了彩色图像RGB通道之间的相关性,因此基于RGB通道联合相似度估计提出了一种新的彩色图像非局部均值去噪方法。在用非局部均值滤波对彩色图像进行去噪时,首先以目标像素为中心确定其支撑区域,然后根据多通道联合相似度估计确定权重,最后采用逐块滤波的方法对每一层进行滤波。并且针对彩色图像中含有的高斯噪声提出了一种新的噪声参数估计方法。由实验结果可以看出该算法比传统的去噪算法在PSNR和FSIM方面都有提高。因此可以看出在图像去噪过程中考虑三通道之间的相关性是必要的,同时也证明了算法的有效性。     

Seabed Segmentation Using Optimized Statistics of Sonar Textures

In this paper, we propose and compare two supervised algorithms for the segmentation of textured sonar images, with respect to seafloor types. We characterize seafloors by a set of empirical distributions estimated on texture responses to a set of different filters. Moreover, we introduce a novel similarity measure between sonar textures in this feature space. Our similarity measure is defined as a weighted sum of Kullback–Leibler divergences between texture features. The weight setting is twofold. First, each filter is weighted according to its discrimination power: The computation of these weights are issued from a margin maximization criterion. Second, an additional weight, evaluated as an angular distance between the incidence angles of the compared texture samples, is considered to take into account sonar-image acquisition process that leads to a variability of the backscattered value and of the texture aspect with the incidence-angle range. A Bayesian framework is used in the first algorithm where the conditional likelihoods are expressed using the proposed similarity measure between local pixel statistics and the seafloor prototype statistics. The second method is based on a variational framework as the minimization of a region-based functional that involves the similarity between global-region texture-based statistics and the predefined prototypes.      

Fast Non-Local Means (NLM) Computation With Probabilistic Early Termination

A speed up technique for the non-local means (NLM) image denoising algorithm based on probabilistic early termination (PET) is proposed. A significant amount of computation in the NLM scheme is dedicated to the distortion calculation between pixel neighborhoods. The proposed PET scheme adopts a probability model to achieve early termination. Specifically, the distortion computation can be terminated and the corresponding contributing pixel can be rejected earlier, if the expected distortion value is too high to be of significance in weighted averaging. Performance comparative with several fast NLM schemes is provided to demonstrate the effectiveness of the proposed algorithm.      

Image denoising based on non-local means filter and its method noise thresholding

Non-local means filter uses all the possible self-predictions and self-similarities the image can provide to determine the pixel weights for filtering the noisy image, with the assumption that the image contains an extensive amount of self-similarity. As the pixels are highly correlated and the noise is typically independently and identically distributed, averaging of these pixels results in noise suppression thereby yielding a pixel that is similar to its original value. The non-local means filter removes the noise and cleans the edges without losing too many fine structure and details. But as the noise increases, the performance of non-local means filter deteriorates and the denoised image suffers from blurring and loss of image details. This is because the similar local patches used to find the pixel weights contains noisy pixels. In this paper, the blend of non-local means filter and its method noise thresholding using wavelets is proposed for better image denoising. The performance of the proposed method is compared with wavelet thresholding, bilateral filter, non-local means filter and multi-resolution bilateral filter. It is found that performance of proposed method is superior to wavelet thresholding, bilateral filter and non-local means filter and superior/akin to multi-resolution bilateral filter in terms of method noise, visual quality, PSNR and Image Quality Index.     

Adaptive segmentation of textured images by using the coupled Markov random field model.

Although simple and efficient, traditional feature-based texture segmentation methods usually suffer from the intrinsical less inaccuracy, which is mainly caused by the oversimplified assumption that each textured subimage used to estimate a feature is homogeneous. To solve this problem, an adaptive segmentation algorithm based on the coupled Markov random field (CMRF) model is proposed in this paper. The CMRF model has two mutually dependent components: one models the observed image to estimate features, and the other models the labeling to achieve segmentation. When calculating the feature of each pixel, the homogeneity of the subimage is ensured by using only the pixels currently labeled as the same pattern. With the acquired features, the labeling is obtained through solving a maximum a posteriori problem. In our adaptive approach, the feature set and the labeling are mutually dependent on each other, and therefore are alternately optimized by using a simulated annealing scheme. With the gradual improvement of features' accuracy, the labeling is able to locate the exact boundary of each texture pattern adaptively. The proposed algorithm is compared with a simple MRF model based method in segmentation of Brodatz texture mosaics and real scene images. The satisfying experimental results demonstrate that the proposed approach can differentiate textured images more accurately.     

基于L1-L2范数的正则项去噪模型的PCB图像去噪算法研究

电子行业常通过提取图像特征来对印刷电路板(Printed Circuit Board,PCB)进行缺陷识别。为了改善PCB图像的视觉效果,提升PCB无损检测的准确率,本文提出了一种基于L1-L2范数的正则项去噪模型的PCB图像去噪算法。首先采用非局部均值(Non Local Mean,NLM)滤波算法将提取的图像分解为结构和纹理两个部分,根据结构框架和纹理细节差异化的物理特性,分别使用Lasso回归算法和Ridge回归算法进行图像去噪,然后将Split Bregman迭代框架应用到去噪模型中,最后通过MATLAB软件平台对所提算法进行实验探究,并从视觉角度和去噪效果指标SNR、SSIM等多方面对算法进行评估。实验结果证明了基于L1-L2范数的正则项去噪模型的PCB图像去噪算法的有效性和可行性。     

采用非局部均值的连续太赫兹图像去噪处理

太赫兹辐射能够穿透大多数对可见光和近红外光不透明的物质。提高成像质量是成像系统的关键,尤其对探测器性能较低的面阵成像更为重要。通过数字图像处理方法改善成像质量是一条重要的解决途径。应用非局部均值(NLM)分别对真实的透射扫描、反射扫描和透射面阵、反射面阵的太赫兹图像进行了去噪处理,选取不同的控制参数进行了对比分析,同时对比了均值滤波处理结果。实验结果表明:非局部均值能够较好地去除连续太赫兹图像噪声、提高成像质量,对噪声严重的面阵成像去噪效果最明显。非局部均值去噪并保持边缘能力明显好于均值滤波。     

基于块匹配和三维滤波的相干激光雷达图像去噪研究

针对相干激光雷达距离像和强度像的不同噪声特性,分别采用基于块匹配和三维滤波(BM3D)算法对距离像和强度像进行了去噪处理,并与非局部均值(NLM)算法的去噪结果进行了比较和分析。实验仿真结果表明,对于强度像,同态BM3D算法明显好于同态NLM算法,在保持目标区域的灰度分布均匀性、保持及恢复图像边缘等方面具有较好的性能;对于距离像,由于一般更注重保持正确的距离值,BM3D和NLM算法均不够理想。     

基于统计假设测试的图像结构检测方法及修复

图像去噪是数字图像处理中的一个重要方法,任何图像去噪算法都不可避免地会把一些边缘或纹理等图像结构给平滑掉,基于修复去噪后的图像中一些图像结构的目的,首先根据统计假设测试原理给出一种度量图像结构特征的方法．通过图像结构特征度可以判断出局部区域是边缘纹理区域还是纯噪声区域,然后应用图像结构特征度来修复被去噪处理后的图像结构...     

一种改进的非局部平均去噪方法

 对非局部平均去噪算法提出了以下改进：首先,利用图像中具有对称结构的性质,在相似性邻域的比较中引入邻域的对称变换,更好地利用了图像的自相似性质;其次,提出一种基于图像灰度分布统计特性的滤波参数选取方法,能够根据不同像素的特点自适应地选取滤波参数;此外,利用非局部平均算法能有效地保护图像结构信息的性质,提出一种两级非局部平均去噪方法.对测试图像去噪的实验结果表明,与原始算法相比,提出的改进方法能够在保护图像结构信息的前提下更有效地去除噪声,峰值信噪比最多可以提高5.9dB, 去噪效果优于BM-3D方法.     

A new adaptive switching filter with directional impulse detection

A novel adaptive switching filter （ASF） based on directional detection is proposed for denoising the images that are highly corrupted by impulse noise. The proposed algorithm employs an efficient noise detection mechanism. It first employs an efficient method to estimate the differences between the current pixel and its neighbors aligned with 28 directions. The current noise pixel is replaced by a median or a mean value within an adaptive filter window with respect to different noise densities. Experimental results show that the proposed approach can not only achieve very low miss-detection ratio and false-alarm ratio even up to high noise corruption, but also preserve the detailed information of an image very well.     

Joint demosaicing and denoising.

The output image of a digital camera is subject to a severe degradation due to noise in the image sensor. This paper proposes a novel technique to combine demosaicing and denoising procedures systematically into a single operation by exploiting their obvious similarities. We first design a filter as if we are optimally estimating a pixel value from a noisy single-color (sensor) image. With additional constraints, we show that the same filter coefficients are appropriate for color filter array interpolation (demosaicing) given noisy sensor data. The proposed technique can combine many existing denoising algorithms with the demosaicing operation. In this paper, a total least squares denoising method is used to demonstrate the concept. The algorithm is tested on color images with pseudorandom noise and on raw sensor data from a real CMOS digital camera that we calibrated. The experimental results confirm that the proposed method suppresses noise (CMOS/CCD image sensor noise model) while effectively interpolating the missing pixel components, demonstrating a significant improvement in image quality when compared to treating demosaicing and denoising problems independently.     

Smoothing low-SNR molecular images via anisotropic median-diffusion

We propose a new anisotropic diffusion filter for denoising low-signal-to-noise molecular images. This filter, which incorporates a median filter into the diffusion steps, is called an anisotropic median-diffusion filter. This hybrid filter achieved much better noise suppression with minimum edge blurring compared with the original anisotropic diffusion filter when it was tested on an image created based on a molecular image model. The universal quality index, proposed in this paper to measure the effectiveness of denoising, suggests that the anisotropic median-diffusion filter can retain adherence to the original image intensities and contrasts better than other filters. In addition, the performance of the filter is less sensitive to the selection of the image gradient threshold during diffusion, thus making automatic image denoising easier than with the original anisotropic diffusion filter. The anisotropic median-diffusion filter also achieved good denoising results on a piecewise-smooth natural image and real Raman molecular images.