Search

In a recent paper, a method called “texture characteristic based morphological component analysis” (TC-MCA) has been proposed to enhance the performances of texture-based image segmentation algorithms: (1) TC-MCA separates texture into multiple pairs of components each representing a different visual characteristic of the texture; (2) then the TC-MCA algorithm manipulates each component and recombines them to produce the enhanced image where textures are more different, so that the texture-based segmentation is improved. This paper proposes a novel method outperforming TC-MCA by: (1) applying dictionaries based on dual-tree complex wavelet transform (DTCWT) to decompose the image into components representing different visual characteristics of texture, and (2) manipulating the dual-tree complex wavelet coefficients of the texture components to enhance each texture component’s own property and producing an texture-difference-enhanced image by recombining the enhanced components. The wavelet-based texture characteristic morphological component analysis (WT-TC-MCA) is applied as the preprocessing step to some state-of-the-art texture-based segmentation algorithms and compared with other texture enhancing methods, including the TC-MCA, with respect to the accuracy, precision-recall curves of the segmentation results. Experiments demonstrate that our method produces better textural difference enhancement effects and improves texture-based segmentation more than the comparators. [ABSTRACT FROM AUTHOR]

• A fine inpainting method of incomplete image based on features fusion and two-steps inpainting (FFTI) is proposed in this paper. • The DMN+ are used to fuse the features of incomplete image to generate optimization map. • The fused features are input into the network with gradient penalty constraints. • The coarse repair graph is further optimized by the idea of coherence of relevant features. In view of the faultiness that the existing image inpainting methods fail to make full use of the complete region to predict the missing region features when the object features are seriously missing, resulting in discontinuous features and fuzzy detail texture of the inpainting results, a fine inpainting method of incomplete image based on features fusion and two-steps inpainting (FFTI) is proposed in this paper. Firstly, the dynamic memory networks (DMN+) are used to fuse the external features and internal features of the incomplete image to generate the incomplete image optimization map. Secondly, a generation countermeasure generative network with gradient penalty constraints is constructed to guide the generator to rough repair the optimized incomplete image and obtain the rough repair map of the target to be repaired. Finally, the coarse repair graph is further optimized by the idea of coherence of relevant features to obtain the final fine repair graph. It is verified by simulation on three image data sets with different complexity, and compared with the existing dominant repair model for visual effect and objective data. The experimental results show that the results of the model repair in this paper are more reasonable in texture structure, better than other models in visual effect and objective data, and the Peak Signal-to-Noise Ratio of the proposed algorithm in the most challenging underwater targe dataset is 27.01, the highest Structural Similarity Index is 0.949. [ABSTRACT FROM AUTHOR]

Dynamic textures (DTs) are moving sequences of natural scenes with some form of temporal regularity such as boiling water, a flag fluttering in the wind. The motion causes continuous changes in the geometry of dynamic textures thus it is difficult to apply traditional vision algorithms to recognize this class of textures. This paper proposes a scheme for modeling and classification of dynamic textures using a local image descriptor which efficiently encodes texture information in a space-time domain. The proposed descriptor extends the well-known spatial texture descriptor, local binary pattern (LBP), to spatio-temporal domain in order to represent the DT by combining appearance feature with the motion. Although, local binary patterns are used extensively in visual recognition applications due to their excellent performance and computational simplicity, but sometimes unable to differentiate different structures properly due to their dependency on center pixel as a threshold. In this paper, a new descriptor based on a global adaptive threshold is employed to compute the structure pattern of local image patch which differentiates various local image structures more efficiently. However, the LBP pattern defines the spatial structure of a local image patch but it does not give information about the contrast of local image patch. We have used Michelson contrast to compute the difference in luminance in the local texture and clubbed with local structure pattern computed using the proposed descriptor. Extensive experiments on dynamic texture databases (Dyntex, Dyntex++ and UCLA) prove the efficiency of the proposed method. [ABSTRACT FROM AUTHOR]

Automatic detection of the pedestrian lane in a scene is an important task in assistive and autonomous navigation. This paper presents a vision-based algorithm for pedestrian lane detection in unstructured scenes, where lanes vary significantly in color, texture, and shape and are not indicated by any painted markers. In the proposed method, a lane appearance model is constructed adaptively from a sample image region, which is identified automatically from the image vanishing point. This paper also introduces a fast and robust vanishing point estimation method based on the color tensor and dominant orientations of color edge pixels. The proposed pedestrian lane detection method is evaluated on a new benchmark dataset that contains images from various indoor and outdoor scenes with different types of unmarked lanes. Experimental results are presented which demonstrate its efficiency and robustness in comparison with several existing methods. [ABSTRACT FROM AUTHOR]

This paper presents a novel double-layer sparse representation (DLSR) approach, for improving both reconstructive and discriminative capabilities of unsupervised dictionary learning. In supervised/unsupervised discriminative dictionary learning, classical approaches usually develop a discriminative term for learning multiple sub-dictionaries, each of which corresponds to one-class training image patches. As such, the image patches for different classes can be discriminated by coefficients of sparse representation, with respect to different sub-dictionaries. However, in the unsupervised scenario, some of the training patches for learning the sub-dictionaries of different clusters are related to more than one cluster. Thus, we propose a DLSR formulation in this paper to impose the first-layer sparsity on the coefficients and the second-layer sparsity on the clusters for each training patch, embedding both the reconstructive (via the first-layer) and discriminative (via the second-layer) capabilities in the learned dictionary. To address the proposed DLSR formulation, a simple yet effective algorithm, called DLSR-OMP, is developed on the basis of the conventional OMP algorithm. Finally, the experiments verify that our approach can improve reconstruction and clustering performance of the learned dictionaries of the conventional approaches. More importantly, the experimental results on texture segmentation show that our approach outperforms other state-of-the-art discriminative dictionary learning approaches in the clustering task. [ABSTRACT FROM AUTHOR]

It is known that the rotations of real-world color textures will vary arbitrarily. This paper presents a novel, simple, yet powerful method for rotation-invariant color texture classification. Firstly, we define a Distance-based Local Binary Pattern (DLBP) descriptor to characterize the color texture. By learning the joint distribution of the rotation-invariant DLBP and color intensity information, we define our Multiple Sub-DLBPs ( MS _ DLBP ) descriptor. The MS _ DLBP features defined in this paper are invariant to rotation. Here, we also compared seven important color spaces in terms of their effectiveness in our proposed MS _ DLBP approach. The experimental results on the Outex and CUReT databases show the defined DLBP descriptor performs better than the existing color LBP descriptors and the proposed MS _ DLBP approach is very robust to rotation invariance and outperforms state-of-the-art texture analysis methods. Also, HSV color space is shown to outperform the other color spaces in many cases. [ABSTRACT FROM AUTHOR]

Object instance detection is a fundamental problem in computer vision and has many applications. Compared with the problem of detecting a texture-rich object, the detection of a texture-less object is more involved because it is usually based on matching the shape of the object with the shape primitives extracted from an image, which is not as discriminative as matching appearance-based local features, such as the SIFT features. The Dominant Orientation Templates (DOT) method proposed by Hinterstoisser et al. is a state-of-the-art method for the detection of texture-less objects and can work in real time. However, it may well generate false detections in a cluttered background. In this paper, we propose a new method which has three contributions. Firstly, it augments the DOT method with a type of illumination insensitive color information. Since color is complementary to shape, the proposed method significantly outperforms the original DOT method in the detection of texture-less object in cluttered scenes. Secondly, we come up with a systematic way based on logistic regression to combine the color and shape matching scores in the proposed method. Finally, we propose a speed-up strategy to work with the proposed method so that it runs even faster than the original DOT method. Extensive experimental results are presented in this paper to compare the proposed method directly with the original DOT method and the LINE-2D method, and indirectly with another two state-of-the-art methods. [ABSTRACT FROM AUTHOR]

• The point cloud (PC) quality assessment problem is converted into the joint quality assessment problem of texture projection map (TPM) and geometry projection map (GPM). • Through the geometric and texture distortion representations in the GPM and TPM inherited from colored PC in 3D space, some perceptual features are mined from these maps, to achieve the purpose of evaluating the visual quality of colored PC. • Handling with both geometry and texture attributes in compression distortion and noise pollution, a multi-distorted Colored PC Database (denoted as CPCD2.0) with subjective scoring is constructed. • Experimental results demonstrate the superiority of the proposed method. Colored point cloud (PC) will inevitably encounter distortion during its acquisition, processing, coding and transmission, which may affect the visual quality of the colored PC. Therefore, it is necessary to design an effective tool for colored PC quality assessment (PCQA). In this paper, considering the mapping relationship of perception between the colored PC and its corresponding projection images, we propose a novel PCQA method based on texture and geometry projection (denoted as TGP-PCQA). The main idea of the proposed TGP-PCQA method is to obtain texture and geometry projection maps from different perspectives for evaluating the colored PC. Specifically, 4D tensor decomposition is used to obtain the combination and difference information between the reference and distorted texture projection maps for mainly characterizing texture distortion of colored PC. Meanwhile, the edge features of the geometry projection map are calculated to measure the global or local geometry distortion. All of the extracted features are combined to predict an overall quality of colored PC. In addition, this paper establishes a multi-distorted colored PC database named CPCD2.0 with compression distortions and Gaussian noise, which orients to the influence of both geometry and texture components in distortion. Experimental results on two open subjective evaluation databases (IRPC and SJTU-PCQA) and the self-built CPCD2.0 database show that the proposed TGP-PCQA method outperforms the state-of-the-art PCQA methods. We are also providing the self-built CPCD2.0 database free of charge at https://github.com/cherry0415/CPCD2.0. [ABSTRACT FROM AUTHOR]

This paper is concerned with the minimum norm partial quadratic eigenvalue assignment problem (MNPQEAP) for vibrating structures via active feedback control. We propose a gradient-based optimization method for solving the MNPQEAP using the receptance measurements, system matrices and a few undesired open-loop eigenvalues with associated eigenvectors. The real form of our method is also derived. Numerical tests show that the proposed methods are effective for solving the MNPQEAP with multiple-input vibration controls. [ABSTRACT FROM AUTHOR]

Abstract Underground flow path (UFP) plays a significant role in illustrating seismic data and revealing stratigraphic structures presented in the data. It is challenging to interactively illustrate the UFP due to the well-known characteristics of seismic data, i.e., noisy, discontinuous, and low resolution. In this paper, we propose a novel interactive visualization approach to illustrate the seismic data, which employs a bit-array based 3D texture to organize different types of interactions. There are three major merits of the organization scheme: It enables to switch different types of interactions flexibly. It allows to perform progressive seed point tracing to get more accurate UFP structures, and supports different stratigraphic display modes and their flexible switching. The feedback from the domain experts suggests that the proposed approach is capable of better revealing the UFP structure and distribution compared with the existing UFP visualization approaches in this field. [ABSTRACT FROM AUTHOR]

Highlights • A brand new topic of text effects transfer to generate fantastic artworks is raised. • A distribution-based prior for text effects is analyzed and mathematically modeled. • Texture local patterns, global distributions and naturalness are jointly considered. • A flexible mechanism to control the texture consistency for words is provided. Abstract In this paper, we explore the problem of fantastic special-effects synthesis for the typography. The main challenge of this problem lies in the model diversities to illustrate varied text effects for different characters. To address this issue, we exploit the key analytics on the high regularity of the texture spatial distribution for text effects to guide the synthesis process. Specifically, we characterize the stylized patches by their normalized positions relative to the text skeleton and the optimal scales to depict their style elements. Our method first estimates these two features and derives their correlation statistically. They are then converted into soft constraints for texture transfer to accomplish adaptive multi-scale texture synthesis and to make style element distribution uniform. It allows our algorithm to produce artistic typography that well consists with both local texture patterns and the global spatial distribution in the source example. Furthermore, stroke similarities are considered to control the varieties of text effects among multiple characters in a word. Experimental results demonstrate the superiority of our distribution-aware method for various text effects over conventional style transfer methods. In addition, we validate the effectiveness of our algorithm with extensive artistic typography library generation and apply our method to a general application of special effects transfer for stroke-based graphics. [ABSTRACT FROM AUTHOR]

In this paper, we propose a novel steganalytic scheme based on local texture pattern (LTP) to detect binary image steganography. We first assess how the expanded LTPs capture embedding distortions exactly. Considering curse of dimensionality when expanding LTPs, we employ Manhattan distance to measure the pixels correlation in a 5 × 5 sized block and select the pixels with closely correlation to remove some LTPs that are not interested. Although the stego image can maintain good visual quality, steganography scheme changes the inter-pixels correlation of binary image. Therefore we utilize totally 8192 LTPs histogram to define a 8192-dimensional steganalytic feature set. Original images and stego images are classified by ensemble classifier. Experimental results show that the proposed steganalytic method can more effectively detect state-of-the-art binary image steganography schemes compared with other steganalytic schemes. [ABSTRACT FROM AUTHOR]

Extracting meaningful structures from images with complicated texture patterns is challenging since it is hard to separate structure from texture with similar scale or intensity contrast. In this paper, we propose a structure-preserving bilateral texture filtering algorithm to smooth texture while preserving salient structures. We design a dual-scale patch toggle scheme. That is, two-scale patches are exploited to represent pixels alternatively, the smaller ones for pixels located at structure edges and the bigger ones for pixels in textural regions, and DASM (Directional Anisotropic Structure Measurement) on each pixel is estimated to select which type of patch for it. Our algorithm is based on the framework of joint bilateral filtering, so it is fast, easy to implement, yet effective for adaptive image smoothing. In particular, our approach outperforms previous methods in terms of preserving small structures. The proposed method achieves excellent results that illustrate its effectiveness and efficiency. [ABSTRACT FROM AUTHOR]

Noise level is an important parameter for the design of video denoising algorithms in video processing applications. However, in practice, the noise level is unknown in most cases, but most existing denoising algorithms simply assume that the noise level is known beforehand, which severely limits their practical use. In this paper, we propose a novel blind video denoising algorithm via block-based optimal noise estimation that adaptively measures noise level by the principal component analysis. The adjacent frame images are searched to construct similar blocks by the block-matching method. The inter-frame differences of these similar blocks are used to estimate the video noise for the impact suppression of video motion, where the video noise is verified to comply with the normal distribution. The weak texture regions are selected by the thresholding function that is deduced based on the normal distribution. In addition, the proposed noise estimation approach is separately combined with several current popular non-blind video denoising methods to verify its superiority. Experimental results demonstrate that the proposed algorithm with low computational complexity not only has better estimation results, but also outperforms the state-of-the-art methods in most cases. [ABSTRACT FROM AUTHOR]

In this paper, the three level descriptions from LTP and the directional features from LDP are combined to form a new local feature descriptor, referred to as local directional ternary pattern (LDTP) for texture classification. LDTP is a framework, which consists in encoding both contrast information and directional pattern features in a compact way based on local derivative variations. To achieve robustness, the proposed operator first computes for each pixel within its 3 × 3 overlapping grayscale image patch, on the one hand, eight directional edge responses using the eight Frei–Chen masks, and on the other hand, central edge response through the 2nd derivative of Gaussian filter to capture more detailed information. This allows producing a more discriminative encoding than several state-of-the art methods based only on intensity information. Then, spatial relationships among the neighboring pixels through the edge responses are exploited independently with the help of both LDP’s and LTP’s concepts to enhance the discrimination capability. Indeed, the implicit utilization of both concepts of LTP and LDP encodes more information in comparison to the existing directional and derivative methods in less space, and simultaneously allows discriminating more textures. Finally, the resultant LDTP pattern is divided into two distinct parts: local directional ternary pattern upper ( LDTP U ) and local directional ternary pattern lower ( LDTP L ), and the final feature descriptor vector is obtained by linear concatenation of both LDTP U and LDTP L histograms. The experiments carried out on nine publicly available texture datasets demonstrated that the proposed LDTP descriptor achieves classification performance, which is competitive or better than several recent and old state-of-the-art LBP variants. Statistical significance of the achieved accuracy improvement by the proposed descriptor has been also demonstrated through the Wilcoxon signed rank test applied on all the tested datasets. [ABSTRACT FROM AUTHOR]

Dynamic texture (DT) in videos is the combination of texture patterns with motion pat-terns, and DT recognition is a key step in many vision-related applications. Owing to the additional challenges arising from the characterization on temporal organizations of texture elements, the recognition on DTs is more dif?cult than that on static textures. In this paper, a DT descriptor for classi?cation is constructed, which examines the stationary irregularities of spatial and temporal distributions of local binary patterns in DT slices and encodes the irregularities by lacunarity-based features. The proposed descriptor has strong robustness to monotonic illumination changes and modest viewpoint changes, as well as strong discriminability in classification. In comparison with histogram-based methods, our approach is capable of encoding spatio-temporal details on the distribution of DT patterns. It also encodes additional details on the layout of DT patterns that recent fractal-based methods ignore. The proposed descriptor was applied to DT classification, and the experimental results show its power on several benchmark datasets. [ABSTRACT FROM AUTHOR]

Depth-Image-Based Rendering (DIBR) is one of the main fundamental techniques for generating new viewpoints in 3D video applications such as multi-viewpoint video (MVV), free viewpoint video (FVV) and virtual reality (VR). Due to the imperfections of color images, depth maps or texture restoration techniques, several types of distortions occur in synthesized views. However, most of related works evaluated the quality of DIBR-synthesized views by only detecting a specific type of distortion, such as stretching, black holes, blurring, etc., which were unable to accurately evaluate the quality of DIBR-synthesized views. In this paper, a new no-reference image quality assessment method is proposed to evaluate the quality of DIBR-synthesized images by combining multi-layer and multi-scale features of images. To be specific, the distortions introduced by different stages of virtual viewpoint synthesis are first analyzed, and then multi-layer and multi-scale features are extracted to estimate the degree of texture and structure distortions. As a result, individual quality scores associated with two types of distortions (e.g., structural distortion and texture distortion) are aggregated to an overall image quality. Experimental results on two publicly available DIBR datasets show that the method has better performance than the state-of-the-art models. Index Terms: image quality assessment, DIBR-synthesized image, distortion correction, BIQA. [ABSTRACT FROM AUTHOR]

Images of outdoor scenes captured in bad weathers are often plagued by the limited visibility and poor contrast, and such degradations are spatially-varying. Differing from most previous dehazing approaches that remove the haze effect in spatial domain and often suffer from the noise problem, this paper presents an efficient multi-scale correlated wavelet approach to solve the image dehazing and denoising problem in the frequency domain. To this end, we have heuristically found a generic regularity in nature images that the haze is typically distributed in the low frequency spectrum of its multi-scale wavelet decomposition. Benefited from this separation, we first propose an open dark channel model (ODCM) to remove the haze effect in the low frequency part. Then, by considering the coefficient relationships between the low frequency and high frequency parts, we employ the soft-thresholding operation to reduce the noise and synchronously utilize the estimated transmission in ODCM to further enhance the texture details in the high frequency parts adaptively. Finally, the haze-free image can be well restored via the wavelet reconstruction of the recovered low frequency part and enhanced high frequency parts correlatively. The proposed approach aims not only to significantly increase the perceptual visibility, but also to preserve more texture details and reduce the noise effect as well. The extensive experiments have shown that the proposed approach yields comparative and even better performance in comparison with the state-of-the-art competing techniques. [ABSTRACT FROM AUTHOR]

Weed detection is a complicated problem which needs several sources of information to be gathered for successful discrimination. In this paper wavelet texture features were examined to verify their potential in weed detection in a sugar beet crop. Successive steps in a discrimination algorithm were designed to determine the wavelet texture features for each image sub-division to be fed to an artificial neural network. Co-occurrence texture features were determined for each multi-resolution image produced by single-level wavelet transform. Image segmentation was based on the decision made by neural network to label each sub-division as weed or main crop. Optimisation of the algorithm was tried by investigating two manners of discrimination of weeds from the main crop. Principal Component Analysis was used to select 14 from the 52 extracted texture features. Results demonstrated that the wavelet texture features were able to effectively discriminate weeds among the crops even when there was significant amount of occlusion and leaves overlapping. [ABSTRACT FROM AUTHOR]

In this paper we focus on gender classification from face images. Despite advances in equipment as well as methods, automatic face image processing for recognition or even just for the extraction of demographics, is still a challenging task in unrestricted scenarios. Our tests are aimed at carrying out an extensive comparison of a feature based approach with two score based ones. When directly using features, we first apply different operators to extract the corresponding feature vectors, and then stack such vectors. These are classified by a SVM-based approach. When using scores, the different operators are applied in a completely separate way, so that each of them produces the corresponding scores. Answers are then either fed to a SVM, or compared pairwise to exploit Likelihood Ratio. The testbeds used for experiments are EGA database, which presents a good balance with respect to demographic features of stored face images, and GROPUS, an increasingly popular benchmark for massive experiments. The obtained performances confirm that feature level fusion achieves an often better classification accuracy. However, it is computationally expensive. We contribute to the research on this topic in three ways: 1) we show that the proposed score level fusion approaches, though less demanding, can achieve results that are comparable to feature level fusion, or even slightly better given that we fuse a particular set of experts; the main advantage over the feature-based approach relying on chained vectors, is that it is not required to evaluate a complex multi-feature distribution and the training process: thanks to the individual training of experts the overall process is more efficient and flexible, since experts can be easily added or discarded from the final architecture; 2) we evaluate the number of uncertain/ambiguous cases, i.e., those that might cause classification errors depending on the classification thresholds used, and show that with our score level fusion these significantly decreases; despite the final rate of correct classifications, this results in a more robust system; 3) we achieve very good results with operators that are not computationally expensive. [ABSTRACT FROM AUTHOR]

Style transfer has been widely applied in the fashion industry for design assistance. This paper focuses on object-level image style transfer for specific texture regions on fashion images. The current methods mostly utilize extra conditional inputs to obtain regions of interest in fashion images, which take extensive time and cost to prepare these data. To address this issue, we propose a two-stage unsupervised approach named TsrNet for clothing texture style transfer between images. Firstly, we construct a network named Mask-GAN to unsupervisedly split out the clothing texture region. Secondly, we improve CycleGAN to interchange the texture on specific target regions of the two images. Specifically, considering the diversity of fashion texture, we construct a multiscale module (Drf-module) with dynamic perceptual fields and design an image gradient-based texture structure loss (GTS) to perform texture transfer. Qualitative and quantitative experimental results show the effectiveness of our algorithm for clothing-specific region style transfer. • A novel two-stage framework is proposed for garments region-specific style transfer. • The Mask-GAN is proposed for target region extraction in a self-supervised manner. • A multi-scale module with dynamic receptive field is built to generate various styles. • A texture structure loss is built to make the generated styles satisfy human vision. [ABSTRACT FROM AUTHOR]

Schemes to complement context relationships by cross-scale feature fusion have appeared in many RGB-D scene parsing algorithms; however, most of these works conduct multi-scale information interaction after multi-modal feature fusion, which ignores the information loss of the two modes in the original coding. Therefore, a cross-complementary fusion network (CCFNet) is designed in this paper to calibrate the multi-modal information before feature fusion, so as to improve the feature quality of each mode and the information complementarity ability of RGB and the depth map. First, we divided the features into low, middle, and high levels, among which the low-level features contain the global details of the image and the main learning features include texture, edge, and other features. The middle layer features contain not only some global detail features but also some local semantic features. Additionally, the high-level features contain rich local semantic features. Then, the feature information lost in the coding process of low and middle level features is supplemented and extracted through the designed cross feature enhancement module, and the high-level features are extracted through the feature enhancement module. In addition, the cross-modal fusion module is designed to integrate multi-modal features of different levels. The experimental results verify that the proposed CCFNet achieves excellent performance on the RGB-D scene parsing dataset containing clothing images, and the generalization ability of the model is verified by the dataset NYU Depth V2. [ABSTRACT FROM AUTHOR]

In this paper, we propose an end-to-end Attention-adaptive Multi-scale Feature Aggregation Dehazing Network (AMA-Net). The AMA-Net is based on U-Net and designs with three attention-driven modules, Joint Attention Residual Block (JAB), Joint Attention Feature Aggregation Group (JAAG), and Layer Adaptive Attention Feature Aggregation Module (LAA). To be more specific, considering the unevenly distributed haze in images, we introduce the JAB, which adaptively assigns weights to make networks pay attention to important features; to fully utilize the residual features, we propose the residual aggregation (via three JABs) in JAAG; since most feature aggregation methods for dehazing networks do not filter and refine features at different layers, we add LAA to the decoder to weight the features at different layers for aggregation. Through the ablation studies, we verify the effectiveness of the JAB, JAAG, and LAA. Experimental results on synthetic and real-world datasets show that the proposed AMA-Net outperforms relevant state-of-the-art methods. • We propose an Attention-adaptive Multi-scale Feature Aggregation Dehazing Network (AMA-Net). The AMA-Net is based on U-Net and designed with Joint Attention Residual Block (JAB), Joint Attention Feature Aggregation Group (JAAG), and Layer Adaptive Attention Feature Aggregation Module (LAA). • Considering the unevenly distributed haze in images, we introduce the JAB, which adaptively assigns weights to make networks focus on important features. • We propose the JAAG to integrate the residual features of different JABs using novel residual aggregation, which preserves the residual features containing high-frequency texture information during dehazing. • We establish the LAA to weight and aggregate the features at different layers in the decoder to consider the relevance of the features. This operation enables image feature information in the intermediate layers to be filtered and utilized. [ABSTRACT FROM AUTHOR]

In High Efficiency Video Coding (HEVC), intra coding plays an important role, but also involves huge computational complexity due to a flexible coding unit (CU) structure and a large number of prediction modes. This paper presents a fast algorithm based on the sole- and multi-depth texture measurements to reduce the complexity from CU size and prediction mode decisions. For the CU size decision, evaluation results in the CU coding with one and multiple depths are utilized to classify CUs into heterogeneous, homogeneous, depth-prominent and other ones. Fast CU size decisions are made for different kinds of CUs. For the prediction mode decision, the tendencies for different CU sizes are detected based on multiple depths. The number of searching modes is decreased adaptively for the CU size with fewer tendencies. Experimental results show the proposed algorithm by off-line training reduces 53.32% computational complexity, with 1.47% bit-rate increasing. [ABSTRACT FROM AUTHOR]

With the great success in three-dimensional (3D) movies, a lot of 3D content have been generated. Depth information is one of the important elements in 3D content generation. Stereo matching methods obtain depth information using the characteristic of binocular disparity. These methods find corresponding points between two images which have different viewpoints to calculate the disparity value. However, these methods have difficulties computing accurate disparity values in the textureless region. Smeared pixels near the edge region also make difficult for the stereo matching. In this paper, we propose a pixel based cost computation for the cross-scale stereo matching using the distance transform to improve these problems. In addition, the disparity error detection and correction methods are also proposed as a post-processing step. As a result, we obtain the enhanced disparity map which is robust to the textureless region and the edge region. [ABSTRACT FROM AUTHOR]

It is difficult for us to generate a gradually changed texture by combing two textures when they have very different textures and structures. This is a common problem in constructing panoramas and in pasting from one texture to another. The inconsistency between the sources is what we want to deal with in this paper. We present a new method for synthesizing a transition region between two source textures, such that inconsistent textures and structural properties all change gradually from one source to the other. We first extract the convolutional neural network (CNN) features of two source textures and one initial target texture at convolution and pooling layers. We set the distortion function to be the square of the difference between feature maps of source texture and target texture. And we based on feature maps compute the Gram matrix which is the inner product between feature maps in each layer. We set the second distortion function to be the square of the difference between Gram matrix of second source texture and target texture. Our target function is the weighted sum of two distortion functions we described before. We use Large-scale bound-constrained optimization method to optimize the target function and get the ultimate result. The model provides a new tool to generate a transition region between two source textures by CNN. Our method is robust to various types of images tested. [ABSTRACT FROM AUTHOR]

This paper presents a new depth image recovery method for RGB-D sensors giving a complete, sharp, and accurate object shape from a noisy boundary depth map. The proposed method uses the image guided Total Generalized Variation (TGV) with the depth confidence. A new directional hole filling method of view synthesis is also investigated to produce natural texture in hole regions whereas reducing blurring effect and preventing distortion. Thus, a high-quality image view can be achieved. Experimental results show that the proposed method yields higher quality recovered depth maps and synthesized image views than other previous methods. [ABSTRACT FROM AUTHOR]

The Chan–Vese (C–V) model is an ineffective method for processing images in which the intensity is inhomogeneous. This is especially true for multi-object segmentation, in which the target may be missed or excessively segmented. In addition, for images with rich texture information, the processing speed of the C–V is slow. To overcome these problems, this paper proposes an effective multi-object C–V segmentation model based on region division and gradient guide. First, a rapid initial contour search is conducted using Otsu’s method. This contour line becomes the initial contour for our multi-object segmentation C–V model based on a gradient guide. To achieve the multi-object segmentation the image is then converted to a single level set whose evolution is controlled using an adaptive gradient. The feasibility of the proposed model is analyzed theoretically, and a number of simulation experiments are conducted to validate its effectiveness. [ABSTRACT FROM AUTHOR]

Frequent in practice spatially variant contrast/offset deviations that preserve image appearance hinder classification based on signal co-occurrence statistics. Contrast/offset-invariant descriptors of ordinal signal relations, such as local binary or ternary patterns (LBP/LTP), are popular means to overcome this drawback. This paper extends conventional LBP/LTP-based classifiers towards learning, rather than prescribing most characteristic shapes, sizes, and numbers of these patterns for semi-supervised texture classification and retrieval. The goal is to discriminate a particular texture represented by a single training or query sample from other types of textures. The proposed learning framework models images as samples from a high-order ordinal Markov–Gibbs random field (MGRF). Approximate analytical estimates of the model parameters guide selecting characteristic patterns of a given order, the higher order patterns being learned on the basis of the already found lower order ones. Comparative experiments on four texture databases confirmed that classifiers with the learned multiple LTPs from the 3rd to 8th order consistently outperform more conventional ones with the prescribed 9th-order fixed-shape LBP/LTPs or a few other filters. [ABSTRACT FROM AUTHOR]

Attributes of objects such as “square”, “metallic” , and “red” allow a way for humans to explain or discriminate object categories. These attributes also provide a useful intermediate representation for object recognition, including support for zero-shot learning from textual descriptions of object appearance. However, manual selection of relevant attributes among thousands of potential candidates is labor intensive. Hence, there is increasing interest in mining attributes for object recognition. In this paper, we introduce two novel techniques for nominating attributes and a method for assessing the suitability of candidate attributes for object recognition. The first technique for attribute nomination estimates attribute qualities based on their ability to discriminate objects at multiple levels of the taxonomy. The second technique leverages the linguistic concept of distributional similarity to further refine the estimated qualities. Attribute nomination is followed by our attribute assessment procedure, which assesses the quality of the candidate attributes based on their performance in object recognition. Our evaluations demonstrate that both taxonomy and distributional similarity serve as useful sources of information for attribute nomination, and our methods can effectively exploit them. We use the mined attributes in supervised and zero-shot learning settings to show the utility of the selected attributes in object recognition. Our experimental results show that in the supervised case we can improve on a state of the art classifier while in the zero-shot scenario we make accurate predictions outperforming previous automated techniques. [ABSTRACT FROM AUTHOR]

One challenging research problem of hand pose recognition is the accurate detection of finger abduction and flexion with a single camera. The detection of flexion movements from a 2D image is difficult, because it involves estimation of finger movements along the optical axis of the camera ( z direction). In this paper, a novel approach to hand pose recognition is proposed. We use the concept of object-based video abstraction for segmenting the frames into video object planes (VOPs), as used in MPEG-4, with each VOP corresponding to one semantically meaningful hand position. Subsequently, a particular hand pose is recognized by analyzing the key geometrical features and the textures of the hand. The abduction and adduction movements of the fingers are analyzed by considering a skeletal model. Probabilistic distributions of the geometric features are considered for modeling intra-class abduction and adduction variations. Additionally, gestures differing in flexion positions of the fingers are classified by texture analysis using homogeneous texture descriptors (HTD). Finally, hand poses are classified based on proximity measurement by considering the intra-class abduction and adduction and/or inter-class flexion variations. Experimental results show the efficacy of our proposed hand pose recognition system. The system achieved a 99% recognition rate for one-hand poses and a 97% recognition rate for two-hand poses. [ABSTRACT FROM AUTHOR]

In this paper, we propose a novel image retrieval method called hybrid information descriptors (HIDs) consisting of mutual information descriptors (MIDs) and self information descriptors (SIDs). Based on the physiological structure of human eyes and visual perception mechanism, HIDs are designed to explore the internal correlations among different image feature spaces with image structure and multi-scale analysis, not only characterizing the low-level features, such as color, shape and texture, but also imitating the process of visual information transfer and perception in high-level understanding with the help of the proposed visual optimization model for feature fusion. Comparing with other existing methods applied to content-based image retrieval (CBIR) on four datasets, the usefulness and effectiveness of the HIDs are shown. Extensive experimental results can also demonstrate this. [ABSTRACT FROM AUTHOR]

In this paper, we propose a novel color texture classification method based on statistical characterization. The approach consists in modeling complex wavelet coefficients of both luminance and chrominance components separately leading to a multi-modeling approach. The copula theory allows to take into account the spatial dependencies which exist within the intra-luminance sub-bands via the luminance model M L , and also between the inter-chrominance subband coefficients via the chrominance model M C r . The multi-model, i.e. M L and M C r , is used to develop a Bayesian classifier based on the softmax principal. To derive the classifier, we propose a closed-form expression for the Rao geodesic distance between two copulas. Experiments on two sub-families of luminance-chrominance color spaces namely Lab and HSV have been carried out for a wide range of color texture databases. The combination of different statistical sub-models show that the multi-modeling performs better than some existing methods in term of classification rates. [ABSTRACT FROM AUTHOR]

In traditional image retrieval, images are commonly represented using Bag-of-visual-Words (BoW) built from image local features. However, the lack of spatial and structural information suppresses its performance in applications. In this paper, we introduce a multi-cues description by fusing structural, content and spatial information for partial-duplicate image retrieval. Firstly, we propose a rotation-invariant Local Self-Similarity Descriptor (LSSD), which captures the internal structural layouts in the local textural self-similar regions around interest points. Then, based on the spatial pyramid model, we make use of both LSSD and SIFT to construct an image representation with multi-cues. Finally, we formulate the Semi-Relative Entropy as the distance metric. Comparison experiments with state-of-the-art methods on four popular databases show the efficiency and effectiveness of our approach. [ABSTRACT FROM AUTHOR]

With the widespread proliferation of computers, many human activities entail the use of automatic image analysis. The basic features used for image analysis include color, texture, and shape. In this paper, we propose a new shape description method, called Hough Transform Statistics (HTS), which uses statistics from the Hough space to characterize the shape of objects or regions in digital images. A modified version of this method, called Hough Transform Statistics neighborhood (HTSn), is also presented. Experiments carried out on three popular public image databases showed that the HTS and HTSn descriptors are robust, since they presented precision-recall results much better than several other well-known shape description methods. When compared to Beam Angle Statistics (BAS) method, a shape description method that inspired their development, both the HTS and the HTSn methods presented inferior results regarding the precision-recall criterion, but superior results in the processing time and multiscale separability criteria. The linear complexity of the HTS and the HTSn algorithms, in contrast to BAS, make them more appropriate for shape analysis in high-resolution image retrieval tasks when very large databases are used, which are very common nowadays. [ABSTRACT FROM AUTHOR]

Abstract: Patch-based de-noising algorithms and patch manifold smoothing have emerged as efficient de-noising methods. This paper provides a new insight on these methods. We show how to extend them to separate oscillatory patterns that could be entangled. A collection of particular patches, that we call reference set, is selected by the user. We define a notion of similarity relative to this reference set that is used to extend the Non-Local Means (see Buades et al., 2005) [1] and the graph-based de-noising method (see Szlam et al., 2008) [12]. [Copyright &y& Elsevier]

Abstract: The text associated with images provides valuable semantic meanings about image content that can hardly be described by low-level visual features. In this paper, we propose a novel multimodal approach to automatically predict the visual concepts of images through an effective fusion of textual features along with visual ones. In contrast to the classical Bag-of-Words approach which simply relies on term frequencies, we propose a novel textual descriptor, namely the Histogram of Textual Concepts (HTC), which accounts for the relatedness of semantic concepts in accumulating the contributions of words from the image caption toward a dictionary. In addition to the popular SIFT-like features, we also evaluate a set of mid-level visual features, aiming at characterizing the harmony, dynamism and aesthetic quality of visual content, in relationship with affective concepts. Finally, a novel selective weighted late fusion (SWLF) scheme is proposed to automatically select and weight the scores from the best features according to the concept to be classified. This scheme proves particularly useful for the image annotation task with a multi-label scenario. Extensive experiments were carried out on the MIR FLICKR image collection within the ImageCLEF 2011 photo annotation challenge. Our best model, which is a late fusion of textual and visual features, achieved a MiAP (Mean interpolated Average Precision) of 43.69% and ranked 2nd out of 79 runs. We also provide comprehensive analysis of the experimental results and give some insights for future improvements. [Copyright &y& Elsevier]

Abstract: It has been effectively proved that histogram of image fractal coding parameters can be used for image retrieval. In recent years, many researchers have paid more and more attention to this application of image fractal coding. In this paper, a new statistical method, based on kernel density estimation, is used for analyzing fractal coding parameters. The fractal signatures are then extracted for texture image retrieval. Experimental results show that the proposed method not only has higher retrieval rate but also faster retrieval speed than existing method. [Copyright &y& Elsevier]

Abstract: In this paper, structure elements’ descriptor (SED) – a novel texture descriptor, is proposed. SED can effectively describe images and represent image local features. Moreover, SED can extract and describe color and texture features. The image structure elements’ histogram (SEH) is computed by SED, and HSV color space is used (it has been quantized to 72 bins). SEH integrates the advantages of both statistical and structural texture description methods, and it can represent the spatial correlation of color and texture. The results demonstrate that the method has a better performance than other image retrieval methods in the experiments. [Copyright &y& Elsevier]

Abstract: The two principal morphological texture descriptors, granulometry and morphological covariance, rely on the common principle of successive filtering of an image using a variety of structuring elements, from which feature vectors are subsequently computed. A crucial stage of their computation is the numerical characterization or parameterization of each of the filtered images. In this regard, the zero-th statistical moment is the traditional measure, while the use of higher order moments has also been reported. In this paper, we present the results of a comparative study, concentrating on the potential of various statistical moments for the task of parameterization, while additionally investigating the contribution of Fourier transform moments. The experiments are conducted with focus on texture description effectiveness and on noise robustness, using publicly available texture collections: Outex, CUReT and KTH-TIPS2b, where it is shown that the combination of moments leads to superior classification performance even at high noise levels. [Copyright &y& Elsevier]

This paper presents solid waste bin level detection and classification using gray level co-occurrence matrix (GLCM) feature extraction methods. GLCM parameters, such as displacement, d, quantization, G, and the number of textural features, are investigated to determine the best parameter values of the bin images. The parameter values and number of texture features are used to form the GLCM database. The most appropriate features collected from the GLCM are then used as inputs to the multi-layer perceptron (MLP) and the K-nearest neighbor (KNN) classifiers for bin image classification and grading. The classification and grading performance for DB1, DB2 and DB3 features were selected with both MLP and KNN classifiers. The results demonstrated that the KNN classifier, at KNN = 3, d = 1 and maximum G values, performs better than using the MLP classifier with the same database. Based on the results, this method has the potential to be used in solid waste bin level classification and grading to provide a robust solution for solid waste bin level detection, monitoring and management. [Copyright &y& Elsevier]

Although colorful information in natural scenes can be collected, due to the limitation of camera depth of field, it is hard to capture an image with all-in-focus. Sparse representation (SR)-based methods have shown their powerful potentiality and ability in multi-focus image fusion. However, because of sparse coding and information compress, the existing fusion methods based on SR are imperfect to seize the rich details and significant texture information in source images. As a result, a fusion method based on multi-scale sparse representation for registered multi-focus images (MIF-MsSR) is proposed in this paper, where an adaptive fusion rule for sparse coefficients is presented. At first, source images are processed by multi-scale decomposition and sub-images with different scales can be obtained. According to image features with different richness in these sub-images, dictionaries with different sizes and redundancy are thereby trained. By comprehensively considering the relationships of focused areas, out-of-focused areas and boundary areas between the source images, an adaptive fusion rule based on l 0 − max and Sum Modified Laplacian (SML) is proposed. Finally, a fused image with all-in-focus can be obtained by sparse reconstruction and inverse multi-scale decomposition. Excessive experiments on multi-focus images have demonstrated that the proposed MIF-MsSR not only reserves the integrity of the information in source images, but also has better fusion performance on subjective and objective indicators than other state-of-the-art methods. • According to the scale characteristics of multi-scale decomposition, dictionaries with different size and redundancy are trained. • The proposed method employs the different fusion rules to different relationships of focused and out-of-focused areas. • An adaptive fusion rule based on the focus relationships of areas in multi-focus source images is proposed. [ABSTRACT FROM AUTHOR]