Search

Sequence similarity analysis is a basic method in computational biological studies. Determining the similarity of biological sequences is a vital step in much research, such as exploring the evolutionary relationship among species, gene function analysis, protein structure prediction, and sequence retrieving. This paper introduces a method that uses the theory of the gray-level co-occurrence matrix, which is important in image texture analysis, to define and calculate the features of a DNA sequence. The proposed method can make a quantitative analysis and compute the defined texture features of a DNA sequence. Using these quantified sequence features, a similarity distance matrix can be computed and phylogenetic relationships also can be inferred. From the quantified features, we found that the DNA sequence of humans has the highest entropy and lowest energy. From human to chimpanzee, orangutan, gorilla, and other species, the entropy decreases and energy increases. The advantage of the proposed method is that it can compute multiple features inherent in each sequence. Furthermore, the defined features can be the key values or tags for each sequence for sequence retrieval and similarity analysis.

In [1] we developed a new uncertainty measure which incorporates Renyi entropy instead of Shannon entropy. This new uncertainty measure was conjectured to be invariant to the Renyi order α > 0, whereas for discrete signals other than picket fence signal the uncertainty measure decreases for α > 0. In this paper, we prove this invariance, and test whether this invariance is predictive in the problem of texture classification for digital images. In the preliminary results, we find that it does, in that the recognition performance does not depend significantly on the Renyi parameter α. We hope that these results will be extended to other problems where Renyi entropy is used.

The statistical properties of Synthetic Aperture Radar (SAR) image texture reveal useful target characteristics. It is well-known that these images are affected by speckle and prone to extreme values due to double bounce and corner reflectors. The $\mathcal {G}_{I}^{0}$ distribution is flexible enough to model different degrees of texture in speckled data. It is indexed by three parameters: $\alpha $ , related to the texture, $\gamma $ , a scale parameter, and $L$ , the number of looks. Quality estimation of $\alpha $ is essential due to its immediate interpretability. In this letter, we exploit the connection between the $\mathcal G_{I}^{0}$ and Pareto distributions. With this, we obtain six estimators that have not been previously used in the SAR literature. We compare their behavior with others in the noisiest case for monopolarized intensity data, namely single look case. We evaluate them using Monte Carlo methods for noncontaminated and contaminated data, considering convergence rate, bias, mean squared error, and computational time. We conclude that two of these estimators based on the Pareto law are the safest choices when dealing with actual data and small samples, as is the case of despeckling techniques and segmentation, to name just two applications. We verify the results with an actual SAR image.

In this study, we discuss mainly the image denoising and texture retention issues. Usually, the time-fractional derivative has an adjustable fractional order to control the diffusion process, and its memory effect can nicely retain the image texture when it is applied to image denoising. Therefore, we design a new Rudin-Osher-Fatemi model with a time-fractional derivative based on a classical one, where the discretization in space is based on the integer-order difference scheme and the discretization in time is the approximation of the Caputo derivative (i.e., Caputo-like difference is applied to discretize the Caputo derivative). Stability and convergence of such an explicit scheme are analyzed in detail. We prove that the numerical solution to the new model converges to the exact solution with the order of O(τ2−α+h2), where τ, α, and h are the time step size, fractional order, and space step size, respectively. Finally, various evaluation criteria including the signal-to-noise ratio, feature similarity, and histogram recovery degree are used to evaluate the performance of our new model. Numerical test results show that our improved model has more powerful denoising and texture retention ability than existing ones.

In this study, the authors propose a fractional derivative-based image decomposition and denoising model which decomposes the image into the cartoon component (the component formed by homogeneous regions and with sharp boundaries) and the texture (or noise) component. The cartoon component is modelled by a function of the fractional-order total bounded variation, while the texture component is modelled by an oscillatory function, bounded in the negative Sobolev space norm. The authors give the corresponding minimisation functional, after some transformations, and then the resulting fractional-order partial differential equation can be solved using the Fourier transform. By symmetry and asymmetry of the fractional-order derivative, some generalisations and variants of the proposed model are also introduced. Finally, the authors implement the algorithm by the fractional-order finite difference in the frequency-domain. The experimental results demonstrate that the proposed models make objective and visual improvements compared with other standard approaches in the task of decomposition and denoising.

This paper shows the approach how to choose the input parameters for DIC method. An error of a measurement depends on template subset size, search region size texture quality. The influence of the first two parameters is demonstrated on the example of material with homogeneous deformation. The view of displacement vector fields is analyzed and standard errors for displacements and deformation are calculated. Two thresholds for input parameters are found, where minimal subset size is related to an image texture quality, while the second threshold is defined by the requirement that the subset must fall within the search area of the deformed image. The influence of fractal textures of artificial images and its noise on subset size is demonstrated by modeling Monte-Carlo method. Fractal dimension is proposed to characterize the quality of an image texture. Formulas and analysis are given to diminish the error of DIC measurements and evaluate the input parameters.

Purpose Image texture is increasingly used to discriminate tissues and lesions in PET/CT. For quantification or in computer-aided diagnosis, textural feature analysis must produce robust and comparable values. Because statistical feature values depend on image count statistics, we investigated in depth the stability of Haralick features values as functions of acquisition duration, and for common image resolutions and reconstructions. Methods A homogeneous cylindrical phantom containing 9.6 kBq/ml Ge-68 was repeatedly imaged on a Siemens Biograph mCT, with acquisition durations ranging from three seconds to three hours. Images with 1.5, 2, and 4 mm isometrically spaced voxels were reconstructed with filtered back-projection (FBP), ordered subset expectation maximization (OSEM), and the Siemens TrueX algorithm. We analysed Haralick features derived from differently quantized (3 to 8-bit) grey level co-occurrence matrices (GLCMs) as functions of exposure E, which we defined as the product of activity concentration in a volume of interest (VOI) and acquisition duration. The VOI was a 50 mm wide cube at the centre of the phantom. Feature stability was defined for df/dE → 0. Results The most stable feature values occurred in low resolution FBPs, whereas some feature values from 1.5 mm TrueX reconstructions ranged over two orders of magnitude. Within the same reconstructions, most feature value-exposure curves reached stable plateaus at similar exposures, regardless of GLCM quantization. With 8-bit GLCM, median time to stability was 16 s and 22 s for FBPs, 18 s and 125 s for OSEM, and 23 s, 45 s, and 76 s for PSF reconstructions, with longer durations for higher resolutions. Stable exposures coincided in OSEM and TrueX reconstructions with image noise distributions converging to a Gaussian. In FBP, the occurrence of stable values coincided the disappearance of negatives image values in the VOI. Conclusions Haralick feature values depend strongly on exposure, but invariance exists within defined domains of exposure. Here, we present an easily replicable procedure to identify said stable exposure domains, where image noise does not substantially add to textural feature values. Only by imaging at predetermined feature-invariant exposure levels and by adjusting exposure to expected activity concentrations, can textural features have a quantitative use in PET/CT. The necessary exposure levels are attainable by modern PET/CT systems in clinical routine.

The widely used energy features extracted from the wavelet domain can effectively represent the common image textures. However, they are not robust to the rotated textures. In this letter, we propose a multiscale rotation-invariant representation (MRIR) of textures by using multiscale sampling. Particularly, a multiscale wavelet transform is used to decompose the magnitude pattern (MP) mapping of a texture. Furthermore, the sign pattern (SP) mapping of a texture is used as a step function, which is further sampled and used to fit the wavelet subbands of the MP mapping for computing the sampled directional mean vectors (SDMVs) of the subbands. Moreover, we construct frequency vectors (FVs) of those SP mappings for capturing the structural information of textures. Finally, we can obtain the MRIR vector of an image texture by concatenating those SDMVs and FVs for texture classification. The comprehensive experimental results demonstrate that our proposed approach outperforms six representative texture classification methods.

Rapid growth in the field of quantitative digital image analysis is paving the way for scientific researchers to make precise measurements about objects in an image. To compute quantities from an image such as the density of compressed materials or the velocity of a shockwave, object boundaries must first be determined. Images containing regions that each have a spatial trend in intensity are of particular interest here. For edge detection, we present a supervised, statistical image segmentation method that incorporates spatial information to locate boundaries between regions with overlapping intensity histograms, specifically for images where the regions are known but precise boundary locations are unknown. The segmentation of a pixel is determined by comparing its intensity to distributions from nearby pixel intensities, and a gradient of the segmented image indicates edge locations. Because of the statistical nature of the algorithm, we use maximum likelihood estimation to quantify uncertainty about ea...

In the existing tolerance granular space models, grid points in each layer a re established only taking space position into account, which ignore uncertainties of image texture, such as randomness, fuzziness and relevance. As a matter of fact, it is very important to extract grid points for constructing tolerance granular spac e, which are tolerance granules' position or center. Therefore, it is very meaningful for the accurate texture feature-description of ima ges to extract grid points well. To address this issue, we firstly apply cloud model to extracting grid points, and establish two new tolerance granular space models. Then, similarity measures based on cloud model and tolerance granule space are presented and two novel image retrieval methods are introduced, including an image texture recognition and a color image retrieval method. Finally, simulation ex periments are done on images of image test set chosen from Corel Database, to compare our proposed methods with the conventional color histogram-based image retrieval method, the salient regions and nonsubsampled contourlet transform-based image retrieval method, and tolerance granular-based multi-level texture image retrieval method. The experimental results demonstrate that the proposed methods are indeed efficient and of practical value to many real-world problems.

The existing active contour models can not achieve accurate segmentation of SAR river images. To solve this difficulty, a novel active contour model driven by median global image fitting energy is proposed. First, the median global fitted image is defined. Then by minimizing the difference between the median global fitted image and the original image, the energy functional of the proposed model is obtained. Moreover, the within-cluster absolute differences of the pixel grayscale values inside and outside the curve are introduced to adaptively adjust the proportions of the region energies inside and outside the curve. Compared with the popular active contour models, extensive experimental results demonstrate the proposed model has clear advantages in terms of both segmentation performance and segmentation efficiency.

The perceptual adaptation of the image (PAI) is introduced by inspiration from Chevreul–Mach Bands (CMB) visual phenomenon. By boosting the CMB assisting illusory effect on boundaries of the regions, PAI adapts the image to the perception of the human visual system and thereof increases the quality of the image. PAI is proposed for application to standard images or the output of any image processing technique. For the implementation of the PAI on the image, an algorithm of morphological filters (MFs) is presented, which geometrically adds the model of CMB effect. Numerical evaluation by improvement ratios of four no-reference image quality assessment (NR-IQA) indexes approves PAI performance where it can be noticeably observed in visual comparisons. Furthermore, PAI is applied as a postprocessing block for classical morphological filtering, weighted morphological filtering, and median morphological filtering in cancelation of salt and pepper, Gaussian, and speckle noise from MRI images, where the above specified NR-IQA indexes validate it. PAI effect on image enhancement is benchmarked upon morphological image sharpening and high-boost filtering.

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 new edge-preserving image smoothing method is proposed.We formulate the problem as an unified edge and texture regularized framework.An efficient solver is proposed to solve the regularized problem.Extensive experiments have been performed to verify the superiority of the work.The proposed smoothing method is employed in many image manipulations.

Thresholding technique is one of the most imperative practices to accomplish image segmentation. In this paper, a novel thresholding algorithm based on 3D Otsu and multi-scale image representation is proposed for medical image segmentation. Considering the high time complexity of 3D Otsu algorithm, an acceleration variant is invented using dimension decomposition rule. In order to reduce the effects of noises and weak edges, multi-scale image representation is brought into the segmentation algorithm. The whole segmentation algorithm is designed as an iteration procedure. In each iteration, the image is segmented by the efficient 3D Otsu, and then it is filtered by a fast local Laplacian filtering to get a smoothed image which will be input into the next iteration. Finally, the segmentation results are pooled to get a final segmentation using majority voting rules. The attractive features of the algorithm are that its segmentation results are stable, it is robust to noises and it holds for both bi-level and multi-level thresholding cases. Experiments on medical MR brain images are conducted to demonstrate the effectiveness of the proposed method. The experimental results indicate that the proposed algorithm is superior to the other multilevel thresholding algorithms consistently.

Local patterns record the gray-level differences between a referenced pixel in an image and its surrounding pixels, which have been commonly used to describe the image features. However, traditional local patterns ignore the spatial distribution feature of texture information in images. We group the gray-level variations along three directions, i.e., horizontal, vertical, and diagonal directions. Each group is then merged into a Local spatial distribution pattern (LSDP) to represent the spatial distribution image feature. We also construct the LSDP patterns for gradient and filtered images, and finally form the Complete local spatial distribution pattern (CLSDP) descriptor to completely describe the texture image feature. Experiments on textural and natural image sets were conducted to compare our CLSDP-based image retrieval algorithm with four previous competitors. The results show that our method is superior to existing algorithms considering both average precision and recall.

In this letter, a no-reference (NR) hybrid image quality assessment (IQA) metric based on cartoon-texture decomposition (CTD) is presented. Focusing on images distorted by both blur and noise, the method takes properties of CTD to separate image into cartoon part with salient edges and texture part with noises. Then, the blur degree and noise level can be estimated separately from different image parts, combine with a joint effect prediction between blur and noise distortions, we present a cartoon-texture decomposition-based blind metric (CTDBBM). Comparative studies with classical full-reference IQA metrics and state-of-the-art NR metrics are conducted on multidistortion image database: LIVEMD. Experimental results show that the CTDBBM performs well and has a high consistency with the human opinions given in the database.

Image denoising can reduce the perturbation inevitably generated during image signal acquisition and its subsequent processing. While the utilization of nonlocal properties can enhance the performance of the state-of-the-art denoising methods, a heavy computational burden is incurred especially for color images. Inspired by the high correlation in the texture information over color channels, for a reduction of the computational burden, this letter proposes denoising the luma channel first, and then, performing a patch-wise linear prediction to transfer the texture information of the denoised luma channel to the other two channels. The texture transferring is adapted to local characteristic (i.e., variance of the local patches) for a reduction of color smearing caused by large prediction error especially along edges. Experimental results confirm that the proposed method achieves performance improvement over the state-of-the-art color image denoising methods only at a slightly increased complexity of single-channel denoising.

Heavy noises distribute in the images when imaging in a poor environment. The randomness of noises makes pixels distributing singularly, which weakens the 1-D piecewise smooth property of original scenes. Thus, wavelets-based compression method no longer works well. In this paper, a layer segmentation based compression scheme is proposed for gray images. Image textures and some high frequency noises are described in a high frequency layer while the coarse part of the image is described in the low frequency layer. The high frequency layer is represented by a joint dictionary, and the low frequency layer is coded with the traditional wavelets. The proposed scheme is tested on nature images and synthetic images. The results show that the proposed scheme achieves better rate-distortion performance compared with several competing compression systems. Besides, further degradation of edges is avoided by the proposed compression scheme.

In order to effectively retrieve a large database of images, a method of creating an image retrieval system CBIR (contentbased image retrieval) is applied based on a binary index which aims to describe features of an image object of interest. This index is called the binary signature and builds input data for the problem of matching similar images. To extract the object of interest, we propose an image segmentation method on the basis of low-level visual features including the color and texture of the image. These features are extracted at each block of the image by the discrete wavelet frame transform and the appropriate color space. On the basis of a segmented image, we create a binary signature to describe the location, color and shape of the objects of interest. In order to match similar images, we provide a similarity measure between the images based on binary signatures. Then, we present a CBIR model which combines a signature graph and a self-organizing map to cluster and store similar images. To illustrate the proposed method, experiments on image databases are reported, including COREL,Wang and MSRDI.

This letter is devoted to the problem of rotation invariant texture classification. Novel rotation invariant feature, symmetric dense microblock difference (SDMD), is proposed which captures the information at different orientations and scales. N-fold symmetry is introduced in the feature design configuration, while retaining the random structure that provides discriminative power. The symmetry is utilized to achieve a rotation invariance. The SDMD is extracted using an image pyramid and encoded by the Fisher vector approach resulting in a descriptor which captures variations at different resolutions without increasing the dimensionality. The proposed image representation is combined with the linear SVM classifier. Extensive experiments are conducted on four texture data sets [Brodatz, UMD, UIUC, and Flickr material data set (FMD)] using standard protocols. The results demonstrate that our approach outperforms the state of the art in texture classification. The MATLAB code is made available.

Exemplar-based texture synthesis is defined as the process of generating, from an input texture sample, new texture images that are perceptually equivalent to the input. In the present work, we model texture self-similarity with conditional Gaussian distributions in the patch space in order to extend the use of stitching techniques. Then, a multiscale texture synthesis algorithm is introduced, where texture patches are modeled at each scale as spatially variable Gaussian vectors in the patch space. The Gaussian distribution for each patch is inferred from the set of its nearest neighbors in the patch space obtained from the input sample. This approach is tested over several real and synthetic texture images, and its results show the effectiveness of the proposed technique for a wide range of textures.

A feature-dependent variational level set formulation is proposed for image segmentation. Two second order directional derivatives act as the external constraint in the level set evolution, with the directional derivative across the image features direction playing a key role in contour extraction and another only slightly contributes. To overcome the local gradient limit, we integrate the information from the maximal (in magnitude) second-order directional derivative into a common variational framework. It naturally encourages the level set function to deform (up or down) in opposite directions on either side of the image edges, and thus automatically generates object contours. An additional benefit of this proposed model is that it does not require manual initial contours, and our method can capture weak objects in noisy or intensity-inhomogeneous images. Experiments on infrared and medical images demonstrate its advantages.

Texture feature extraction methods have been improved greatly in recent years. It is widely known that the local texture feature descriptor can achieve desired performance under the change of image geometric size, different poses and complex illumination conditions. In this paper, a novel local texture descriptor is proposed, named as the multi-degrees improved local difference binary (ILDB). Local difference binary is an promising feature description method, while it only computes the intensity and gradient difference on pairwise grid cells and ignores the image grid inherent texture gradient difference. ILDB can represent difference and texture information of the grid cells intensity and gradient simultaneously. In addition, the multiple-degree strategy is adopted to achieve richer texture description. At the same time, the optimized mutual information is proposed to capture more discriminant feature selection and reduce the dimensionality of the ILDB. Experimental results demonstrate that the proposed method is highly efficient and distinctive compared with several state-of-the-art approaches. Due to good performance of ILDB, it is expected that ILDB has a potential for widespread application in many computer vision fields.

To enhance security of three-dimensional images, an inter-view local texture analysis (ILTA) based stereo image reversible data hiding method is presented. Due to low accuracy of existing predictors, two novel predictors are proposed to improve the prediction precision. In the first predictor, a texture analysis model is built by using ILTA, in which the texture similarity between a pair of matched pixels in the stereo image is used to classify pixels into horizontal texture, vertical texture, smooth and complex types. Thus, the accurate prediction is adaptively computed by considering the pixel type. Moreover, an intra-view based predictor as the second predictor is also described to predict pixels by optimal weights finding (OWF). Since ILTA and OWF predictors are combined to predict pixels in the stereo image, sharp prediction error histograms of two views are both constructed, and then multi-level histogram shifting is used to embed secret data reversibly for obtaining low image distortion and high embedding capacity. Experimental results demonstrates that ILTA and OWF predictors can obtain precise predicted values, and the proposed data hiding method outperforms some state-of-the-art data hiding methods in terms of embedding capacity and quality of stego stereo image.

The method that binarizes an image by local thresholds in separated image areas is useful when the image has uneven brightness. A method was proposed to separate an image dynamically for local thresholding to generate a binary image by using a genetic algorithm GA. However, the existing method used only horizontal and vertical lines for image separation. This paper suggests a method of separating an image dynamically using diagonal lines for local thresholding in order to generate a binary image by using a GA and of evaluating binary images. The experimental results show that the images binarized by the proposed method have good separation of objects from the background, and that they include less noise and blur than those binarized by the existing method.

This letter addresses the texture classification problem through a pixel-based local binary pattern (LBP) statistics aggregation mechanism. Real-world texture images often present challenges for classification algorithms in terms of intra-class variability due, among others, to variable illumination. The LBP operator, a state-of-the-art texture descriptor, possesses key properties for tackling real-world texture images: discriminative power and invariance against monotonic gray level changes. We propose a novel texture classification approach that increases the robustness of LBP-based methods with respect to any type of intra-class variations. The method locally characterizes each pixel with an LBP code histogram and globally computes the label of a textured image by aggregating pixel labels through a voting process. Our approach can be in principle applied to any LBP version, as it focuses on how statistics are computed from LBP codes. We show that the proposed pixel-based approach improves upon traditional LBP block-based approaches in terms of classification accuracy by up to 5.1 p.p. on the public Outex database for the classic LBP with various neighborhoods as well as for various LBP extensions.

We present efficient graph cut algorithms for three problems: (1) finding a region in an image, so that the histogram (or distribution) of an image feature within the region most closely matches a given model; (2) co-segmentation of image pairs and (3) interactive image segmentation with a user-provided bounding box. Each algorithm seeks the optimum of a global cost function based on the Bhattacharyya measure, a convenient alternative to other matching measures such as the Kullback–Leibler divergence. Our functionals are not directly amenable to graph cut optimization as they contain non-linear functions of fractional terms, which make the ensuing optimization problems challenging. We first derive a family of parametric bounds of the Bhattacharyya measure by introducing an auxiliary labeling. Then, we show that these bounds are auxiliary functions of the Bhattacharyya measure, a result which allows us to solve each problem efficiently via graph cuts. We show that the proposed optimization procedures converge within very few graph cut iterations. Comprehensive and various experiments, including quantitative and comparative evaluations over two databases, demonstrate the advantages of the proposed algorithms over related works in regard to optimality, computational load, accuracy and flexibility.

This letter introduces a new image descriptor named Local Topographic Shape Pattern (LTSP) for texture description by relying on complete shape-based image representation. Firstly, a texture image is decomposed into a tree of shapes, i.e., topographic map, to obtain a multi-scale and contrast-invariant representation. Secondly, the shape of each node in the tree is handled with four shape patterns which are proposed as spatial codes of the shape. Finally, statistical histograms of shape patterns are used as texture descriptors. Contrast experiments of the proposed method show satisfactory performance in terms of both texture retrieval and classification tasks when applied to the Brodatz and UMD texture datasets.

We introduce a monotonic sequence based thresholding in image decomposition.We propose of sequence or weighted thresholding based image enhancement.We propose the Fibonacci thresholding in image enhancement.The presented thresholding-based enhancement out-performs competing algorithms. Both image enhancement and image segmentation are important pre-processing steps for various image processing fields including autonomous navigation, remote sensing, computer vision, and biomedical image analysis. Both methods have their merits and their short comings. It then becomes obvious to ask the question: is it possible to develop a new better image enhancement method which has the key elements from both segmentation and image enhancement techniques? The choice of the threshold level is a key task in image segmentation. There are other challenges of image segmentation. For example, it is very difficult to perform the image segmentation in poor data such as shadows and noise. Recently, a homothetic curves Fibonacci-based cross sections thresholding has been developed for the de-noising purposes. Is it possible to develop a new image cross sections thresholding method, which can be used for both segmentation and image enhancement purposes? This paper a) describes a unified approach for signal thresholding, b) extends cross sections concept by generating and using a new class of monotonic, piecewise linear, sequences (slowly or faster growing than Fibonacci numbers) of numbers; c) uses the extended sections concept to the image enhancement and segmentation applications. Extensive experimental evaluation demonstrates that the newly proposed monotonic sequences have great potential in image processing applications, including image segmentation and image enhancement applications. Moreover, study has shown that the generalized cross techniques are invariant under morphological transformations such as erosion, dilation, and median, able to be described analytically, can be implemented by using the look up table methods.

Feature inconsistency of image components is used to obtain tampering detection.The proposed method can obtain accurate tampered regions in composite images.Image-component-based composite image detection is first proposed. This paper proposes an effective composite image detection method that uses the feature inconsistency of image components of the composite image to detect tampered regions. The composite image is first divided into image components. Next, the variance of the noise remaining after de-noising in each image component is calculated and used as a feature. Finally, tampered regions are detected using this feature based on a tampering detection rule. Experimental results show that the proposed method has good composite image detection performance.

In this work we develop a post-processing algorithm which enhances the results of the existing image deblurring methods. It performs additional edge sharpening using grid warping. The idea of the proposed algorithm is to transform the neighborhood of the edge so that the neighboring pixels move closer to the edge, and then resample the image from the warped grid to the original uniform grid. The proposed technique preserves image textures while making the edges sharper. The effectiveness of the method is shown for basic deblurring methods on LIVE database images with added blur and noise.

We propose a novel algorithm to segment a low depth-of-field (DOF) image into its focused region-of-interest (ROI) and defocused background using adaptive second-order statistics (ASOS). Most previous methods depend on the assump -tion that the images are in noise-free conditions, which leads to high false positive rates in noisy images. In this letter, we introduce a novel image segmentation algorithm for noisy low-DOF images. Specifically, we propose a novel feature transform method, called ASOS, which indicates the spatial distribution of the high-frequency components in the face of noisy low-DOF images. Experimental results demonstrate that the proposed method is effective for image segmentation in noisy images compared to several state-of-the-art methods proposed in the literature.

With the advent of powerful ranging and visual sensors, nowadays, it is convenient to collect sparse 3-D point clouds and aligned high-resolution images. Benefitted from such convenience, this letter proposes a joint method to perform both depth assisted object-level image segmentation and image guided depth upsampling. To this end, we formulate these two tasks together as a bi-task labeling problem, defined in a Markov random field. An alternating direction method (ADM) is adopted for the joint inference, solving each sub-problem alternatively. More specifically, the sub-problem of image segmentation is solved by Graph Cuts, which attains discrete object labels efficiently. Depth upsampling is addressed via solving a linear system that recovers continuous depth values. By this joint scheme, robust object segmentation results and high-quality dense depth maps are achieved. The proposed method is applied to the challenging KITTI vision benchmark suite, as well as the Leuven dataset for validation. Comparative experiments show that our method outperforms stand-alone approaches.

The choice of natural image prior decides the quality of restored image. Recently success- ful algorithms exploit heavy-tailed gradient distribution as image prior to restore latent image with piecewise smooth regions. However, these prior assumed also remove the mid-frequency compo- nent such as textural details regions while they preserve sharp edges. That because gradient profile in fractal-like texture do not have sparse characteristic. To restore textural features of expected latent image, in this paper, we introduce fractional-order gradient as image prior which is more appropriate to describe characteristic of image textures. From details comparison of our experiments, the textual details are more clear and visual quality is im- proved.

Image super-resolution using self-optimizing mask via fractional-order gradient interpolation and reconstruction aims to recover detailed information from low-resolution images and reconstruct them into high-resolution images. Due to the limited amount of data and information retrieved from low-resolution images, it is difficult to restore clear, artifact-free images, while still preserving enough structure of the image such as the texture. This paper presents a new single image super-resolution method which is based on adaptive fractional-order gradient interpolation and reconstruction. The interpolated image gradient via optimal fractional-order gradient is first constructed according to the image similarity and afterwards the minimum energy function is employed to reconstruct the final high-resolution image. Fractional-order gradient based interpolation methods provide an additional degree of freedom which helps optimize the implementation quality due to the fact that an extra free parameter $\alpha$-order is being used. The proposed method is able to produce a rich texture detail while still being able to maintain structural similarity even under large zoom conditions. Experimental results show that the proposed method performs better than current single image super-resolution techniques., Comment: 24 pages, 13 figures, it is to appear in ISA Transactions

We extend the theoretical results obtained by Stevenson et al. (1987) to two distributions and provide a quantitative comparison of 1D morphological filter edge localization with two classical types of kernel-based smoothing filters (the mean and the close relative to morphological filters, the median filter). Implications in the context of statistical texture segmentation are briefly discussed.

Image segmentation and image denoising are two important and fundamental topics in the field of image processing. Geometric active contour model based on level set method can deal with the problem of image segmentation, but it does not consider the problem of image denoising. In this paper, a new diffusion equation model for noisy image segmentation is proposed by incorporating some classical diffusion equation denoising models into the segmental process. An assumption about the connection between the image intensity and level set function is given firstly. Some classical denoising models are employed to describe the evolution of level set function secondly. The final nonlinear diffusion equation model for noisy image segmentation is built thirdly. Then image segmentation and image denoising are combined in a united framework. The segmental results can be presented by level set function. Experimental results show that the new model has the advantage of noise resistance and is superior to traditional segmentation model.

There are two major approaches to content-based image retrieval using local image descriptors. One is descriptor-by-descriptor matching and the other is based on comparison of global image representation that describes the set of local descriptors of each image. In large-scale problems, the latter is preferred due to its smaller memory requirements; however, it tends to be inferior to the former in terms of retrieval accuracy. To achieve both low memory cost and high accuracy, we investigate an asymmetric approach in which the probability distribution of local descriptors is modeled for each individual database image while the local descriptors of a query are used as is. We adopt a mixture model of probabilistic principal component analysis. The model parameters constitute a global image representation to be stored in database. Then the likelihood function is employed to compute a matching score between each database image and a query. We also propose an algorithm to encode our image representation into more compact codes. Experimental results demonstrate that our method can represent each database image in less than several hundred bytes achieving higher retrieval accuracy than the state-of-the-art method using Fisher vectors.

This study presents a variational image restoration algorithm based on variable exponential fractional‐order total variation (TV), variable exponential integer‐order TV and tight frame sparse regularization. The energy functional of this variational problem is composed of four parts: a fractional‐order TV regularization term with a variable exponent, an integer‐order TV regularization term with a variable exponent, a data fidelity term and a tight frame regularization term. The variable exponent is a function of the gradient information of the image. The first three parts of the energy functional is the smooth part, and the last part is the non‐smooth part. To minimize the smooth part, the optimization problem can be simply transformed into a gradient descent flow using the variational method. For the non‐smooth part, we use the soft‐thresholding method over the sparse framelet coefficients. As the combination of the fractional‐order derivative and integer‐order derivative, the variable exponent and the sparse regularisation, the proposed method can effectively remove noise of the image, protect the image boundary, and keep image texture details well. Experiments with simulated data and real data are provided to validate the effectiveness of proposed method. This method is robust to noise and is of some practical application value.