Texture Representation Research Articles

Rotation invariant curvelet based image retrieval & classification via Gaussian mixture model and co-occurrence features

Demand for better retrieval methods continue to outstrip the capabilities of available technologies despite the rapid growth of new feature extraction techniques. Extracting discriminatory features that contain texture specific information are of crucial importance in image indexing. This paper presents a novel rotation invariant texture representation model based on the multi-resolution curvelet transform via co-occurrence and Gaussian mixture features for image retrieval and classification. To extract these features, curvelet transform is applied and the coefficients are obtained at each scale and orientation. The Gaussian mixture model (GMM) features are computed from each of the sub bands and co-occurrence features are computed for only specific sub band. Rotation invariance is provided by applying cycle-shift around the GMM features. The proposed method is evaluated on well-known databases such as Brodatz, Outex_TC_00010, Outex_TC_00012, Outex_TC_00012horizon, Outex_TC_00012tl84, Vistex and KTH-TIPS. When the feature vector is analyzed in terms of its size, it is observed that its dimension is smaller than that of the existing rotation-invariant variants and it has a very good performance. Simulation results show a good performance achieved by combining different techniques with the curvelet transform. Proposed method results in high degree of success rate in classification and in precision-recall value for retrieval.

On the generalization of color texture-based face anti-spoofing

Despite the significant attention given to the problem of face spoofing, we still lack generalized presentation attack detection (PAD) methods performing robustly in practical face recognition systems. The existing face anti-spoofing techniques have indeed achieved impressive results when trained and evaluated on the same database (i.e. intra-test protocols). Cross-database experiments have, however, revealed that the performance of the state-of-the-art methods drops drastically as they fail to cope with new attacks scenarios and other operating conditions that have not been seen during training and development phases. So far, even the popular convolutional neural networks (CNN) have failed to derive well-generalizing features for face anti-spoofing. In this work, we explore the effect of different factors, such as acquisition conditions and presentation attack instrument (PAI) variation, on the generalization of color texture-based face anti-spoofing. Our extensive cross-database evaluation of seven color texture-based methods demonstrates that most of the methods are unable to generalize to unseen spoofing attack scenarios. More importantly, the experiments show that some facial color texture representations are more robust to particular PAIs than others. From this observation, we propose a face PAD solution of attack-specific countermeasures based solely on color texture analysis and investigate how well it generalizes under display and print attacks in different conditions. The evaluation of the method combining attack-specific detectors on three benchmark face anti-spoofing databases showed remarkable generalization ability against display attacks while print attacks require still further attention.

Rethinking Text

The way we store, distribute and access textural information has undergone a dramatic change starting by the introduction of movable type around the 1450s. The way, however, we present and perceive written information has not changed much since then. But why is that? Technology has been a key driver in what is now called digital media. It provides a broad variety of possibilities to present written information. Until today, these possibilities stay nearly untouched and current textural representation is taken for granted and unalterable. In this article, the authors argue that, for real progress and innovation, people have to rethink text and to accept textual representations in digital media as an independent and alterable media. The authors summarize different approaches to augment text to foster the discussion and drive further developments.

Local Neighborhood Intensity Pattern–A new texture feature descriptor for image retrieval

In this paper, a new texture descriptor based on the local neighborhood intensity difference is proposed for content based image retrieval (CBIR). For computation of texture features like Local Binary Pattern (LBP), the center pixel in a 3 × 3 window of an image is compared with all the remaining neighbors, one pixel at a time to generate a binary bit pattern. It ignores the effect of the adjacent neighbors of a particular pixel for its binary encoding and also for texture description. The proposed method is based on the concept that neighbors of a particular pixel hold significant amount of texture information that can be considered for efficient texture representation for CBIR. The main impact of utilizing the mutual relationship among adjacent neighbors is that we do not rely on the sign of the intensity difference between central pixel and one of its neighbors (Ii) only, rather we take into account the sign of difference values between Ii and its adjacent neighbors along with the central pixels and same set of neighbors of Ii. This makes our pattern more resistant to illumination changes. Moreover, most of the local patterns including LBP concentrates mainly on the sign information and thus ignores the magnitude. The magnitude information which plays an auxiliary role to supply complementary information of texture descriptor, is integrated in our approach by considering the mean of absolute deviation about each pixel Ii from its adjacent neighbors. Taking this into account, we develop a new texture descriptor, named as Local Neighborhood Intensity Pattern (LNIP) which considers the relative intensity difference between a particular pixel and the center pixel by considering its adjacent neighbors and generate a sign and a magnitude pattern. Finally, the sign pattern (LNIPS) and the magnitude pattern (LNIPM) are concatenated into a single feature descriptor to generate a more effective feature descriptor. The proposed descriptor has been tested for image retrieval on four databases, including three texture image databases - Brodatz texture image database, MIT VisTex database and Salzburg texture database and one face database - AT&T face database. The precision and recall values observed on these databases are compared with some state-of-art local patterns. The proposed method showed a significant improvement over many other existing methods.

A Real Time of an Automatic Finger Vein Recognition System

Finger vein recognition biometric system is one of the most current and accurate biometric technologies. Yet, early implementation of this technology is not widely used in real-time applications. In this work, a finger vein-embedded system based on Rasberry-Pi has been presented. In our process, we use four structural directional elements for smoothing finger veins ROIs. A Top-Hat and Bottom-Hat kernel filters are used to enhance the contrast quality of images. For feature extraction step, we used two approaches for the synthesis of attributes including the geometric and texture representations of venous prints. The first one is a Local Directional Code (LDC) descriptor that characterizes texture and directional information of finger vein print. The Improved Gaussian Matched Filter (IMPGMF) is used to extract the finger vein map that characterised geometric venous information. The proposed vision system presents an Error Equal Rate (EER) lower to 0.02 and Identification Rate (IR) higher to 98.99. Moreover, experimental results show that the designed system is fast enough to run the decision of finger vein verification. Performance results show the efficiency and robustness of our system.

Tactile-to-Visual Cross-Modal Transfer of Texture Categorisation Following Training: An fMRI Study.

We investigated the neural underpinnings of texture categorisation using exemplars that were previously learned either within modalities (visual training and visual test) or across modalities (tactile training and visual test). Previous models of learning suggest a decrease in activation in brain regions that are typically involved in cognitive control during task acquisition, but a concomitant increase in activation in brain regions associated with the representation of the acquired information. In our study, participants were required to learn to categorise fabrics of different textures as either natural or synthetic. Training occurred over several sessions, with each fabric presented either visually or through touch to a participant. Pre- and post-training tests, in which participants categorised visual images only of the fabrics, were conducted during a functional magnetic resonance imaging (fMRI) scan. Consistent with previous research on cognitive processes involved in task acquisition, we found that categorisation training was associated with a decrease in activation in brain regions associated with cognitive systems involved in learning, including the superior parietal cortex, dorsal anterior cingulate cortex (dACC), and the right dorsolateral prefrontal cortex (DLFC). Moreover, these decreases were independent of training modality. In contrast, we found greater activation to visual textures in a region within the left medial occipital cortex (MOC) following training. There was no overall evidence of an effect of training modality in the main analyses, with texture-specific regional changes associated with both within- (visual) and cross- (touch) modal training. However, further analyses suggested that, unlike categorisation performance following within-modal training, crossmodal training was associated with bilateral activation of the MOC. Our results support previous evidence for a multisensory representation of texture within early visual regions of the cortex and provide insight into how multisensory categories are formed in the brain.

Retrieving and Classifying Affective Images via Deep Metric Learning

Affective image understanding has been extensively studied in the last decade since more and more users express emotion via visual contents. While current algorithms based on convolutional neural networks aim to distinguish emotional categories in a discrete label space, the task is inherently ambiguous. This is mainly because emotional labels with the same polarity (i.e., positive or negative) are highly related, which is different from concrete object concepts such as cat, dog and bird. To the best of our knowledge, few methods focus on leveraging such characteristic of emotions for affective image understanding. In this work, we address the problem of understanding affective images via deep metric learning and propose a multi-task deep framework to optimize both retrieval and classification goals. We propose the sentiment constraints adapted from the triplet constraints, which are able to explore the hierarchical relation of emotion labels. We further exploit the sentiment vector as an effective representation to distinguish affective images utilizing the texture representation derived from convolutional layers. Extensive evaluations on four widely-used affective datasets, i.e., Flickr and Instagram, IAPSa, Art Photo, and Abstract Paintings, demonstrate that the proposed algorithm performs favorably against the state-of-the-art methods on both affective image retrieval and classification tasks.

A Feature Descriptor Based on Local Normalized Difference for Real-World Texture Classification

In this paper, we propose a normalized difference vector (NDV) for texture representation. Compared to local-binary-pattern-based descriptors, the proposed NDV takes full advantage of the local difference, and the size can be extended flexibly to cover a large local region. We further employ the bag-of-words model to integrate the local descriptors into a global feature representation of an image. In addition, two strategies are introduced for the proposed NDV to achieve rotation invariance. We test the proposed texture descriptor on benchmark datasets, such as AniTex, VehApp, KTH-TIPS2a, OpenSurface, and Kylberg. Classification results demonstrate the superiority of the proposed descriptor over state-of-the-art methods.

Replicating surface texture: Preliminary testing of molding compound accuracy for surface measurements

The use of surface metrology microscopes to quantify surface texture is a powerful tool for the analysis of archaeological materials. Data collected from these microscopes allows for reliable and reproducible measurements of surface characteristics. However, archaeological materials provide some unique challenges for microscopic analysis, including archaeological objects that cannot leave museums or are too large to observe under a microscope. Because of these challenges, many researchers create molds and casts of an artifact's surface prior to measuring surface texture. The replicate surface is assumed to be an accurate representation of the original surface texture; however, limited testing has been done on the reliability of different molding compounds. This paper uses confocal microscopy to test the fidelity of different molding compounds for surface texture measurements. Results suggest that the objects' original surface texture can pose challenges to the accuracy of molds and the resulting casts, and that the intrinsic texture of the molding material impacts surface measurements of the replicated object.

Dynamic Texture Recognition Using Volume Local Binary Count Patterns With an Application to 2D Face Spoofing Detection

In this paper, a local spatiotemporal descriptor, namely, the volume local binary count (VLBC), is proposed for the representation and recognition of dynamic texture. This descriptor, which is similar in spirit to the volume local binary pattern (VLBP), extracts histograms of thresholded local spatiotemporal volumes using both appearance and motion features to describe dynamic texture. Unlike VLBP using binary encoding, VLBC does not exploit the local structure information and only counts the number of 1s in the thresholded codes. Thus, VLBC can include more neighboring pixels without exponentially increasing the feature dimension as VLBP does. Furthermore, a completed version of VLBC (CVLBC) is also proposed to enhance the performance of dynamic texture recognition with additional information about local contrast and central pixel intensities. The proposed method is not only efficient to compute but also effective for dynamic texture representation. In experiments with three dynamic texture databases, namely, UCLA, DynTex, and DynTex++, the proposed method produces classification rates that are comparable to those produced by the state-of-the-art approaches. In addition to dynamic texture recognition, we propose utilizing CVLBC for 2-D face spoofing detection. As an effective spatiotemporal descriptor, CVLBC can well describe the differences between facial videos of valid users and impostors, thus achieving good performance for face spoofing detection. For comparison with other methods, the proposed method is evaluated on three face antispoofing databases: Print-Attack, Replay-Attack, and CAS Face Antispoofing. The experimental results demonstrate the effectiveness of CVLBC for 2-D face spoofing detection.

Local directional ZigZag pattern: A rotation invariant descriptor for texture classification

Local feature descriptors play a key role in texture classification tasks. However, such traditional descriptors are deficient to capture the edges and orientations information and local intrinsic structure of images. This letter introduces a simple, new, yet powerful rotation invariant texture descriptor named Local Directional ZigZag Pattern (Ldzp) by ZigZag scanning for effective representation of texture. Here at first we compute the directional edge information, so called local directional edge map (Ldem) of a texture image using the Kirsch compass mask in six different directions. Then Local ZigZag Pattern (Lzp) is extracted from all Ldem images. Basically, the Lzp characterizes the spatial ZigZag structure based on the relation between reference pixel and its adjacent neighboring pixels and is insensitive to the illumination changes. Finally, the uniform pattern histograms are computed from all directional Lzp maps which are concatenated to form the final Ldzp descriptor. Extensive experiments on texture classification shows the proposed Ldzp descriptor achieves state-of-the-art performance in terms of average classification accuracy when applied to the large and well-known benchmark Outex database. We have also shown that Ldzp descriptor is equally powerful for human face recognition.

On soil textural classifications and soil-texture-based estimations

Abstract. The soil texture representation with the standard textural fraction triplet sand–silt–clay is commonly used to estimate soil properties. The objective of this work was to test the hypothesis that other fraction sizes in the triplets may provide a better representation of soil texture for estimating some soil parameters. We estimated the cumulative particle size distribution and bulk density from an entropy-based representation of the textural triplet with experimental data for 6240 soil samples. The results supported the hypothesis. For example, simulated distributions were not significantly different from the original ones in 25 and 85 % of cases when the sand–silt–clay and very coarse+coarse + medium sand − fine + very fine sand − silt+clay were used, respectively. When the same standard and modified triplets were used to estimate the average bulk density, the coefficients of determination were 0.001 and 0.967, respectively. Overall, the textural triplet selection appears to be application and data specific.

Texture Classification in Extreme Scale Variations Using GANet.

Research in texture recognition often concentrates on recognizing textures with intraclass variations, such as illumination, rotation, viewpoint, and small-scale changes. In contrast, in real-world applications, a change in scale can have a dramatic impact on texture appearance to the point of changing completely from one texture category to another. As a result, texture variations due to changes in scale are among the hardest to handle. In this paper, we conduct the first study of classifying textures with extreme variations in scale. To address this issue, we first propose and then reduce scale proposals on the basis of dominant texture patterns. Motivated by the challenges posed by this problem, we propose a new GANet network where we use a genetic algorithm to change the filters in the hidden layers during network training in order to promote the learning of more informative semantic texture patterns. Finally, we adopt a Fisher vector pooling of a convolutional neural network filter bank feature encoder for global texture representation. Because extreme scale variations are not necessarily present in most standard texture databases, to support the proposed extreme-scale aspects of texture understanding, we are developing a new dataset, the extreme scale variation textures (ESVaT), to test the performance of our framework. It is demonstrated that the proposed framework significantly outperforms the gold-standard texture features by more than 10% on ESVaT. We also test the performance of our proposed approach on the KTHTIPS2b and OS datasets and a further dataset synthetically derived from Forrest, showing the superior performance compared with the state-of-the-art.

Repulsive-and-attractive local binary gradient contours: New and efficient feature descriptors for texture classification

This paper presents new modeling of local binary patterns for texture representation. Referred to as local binary gradient contours (LBGC), the proposed models are expected to better represent the salient local texture structure. Thanks to the flexibility of repulsive-attractive characteristics, which represent the cornerstone of the proposed descriptors, two distinct LBP-like descriptors are built: repulsive and attractive local binary gradient contours (RLBGC and ALBGC). Conventional methods such as LBP, the family of binary gradient contours (BGC1, BGC2 and BGC3), LBP by neighborhoods (nLBPd) and several other LBP-like methods, are based on pairwise comparison of adjacent pixels. Unlike these methods, the proposed RLBGC and ALBGC operators encode the differences between local intensity values within triplets of pixels, along a closed path around the central pixel of a 3x3 gray-scale image patch. In order to increase the robustness of the proposed RLBGC and ALBGC descriptors, the triplet formed by the average local and average global gray levels (ALGL and AGGL) and the central pixel is incorporated in the modeling. In order to make the proposed approach more robust and stable, the RLBGC and ALBGC are concatenated together to form multi-scale repulsive-and-attractive local binary gradient contour (RALBGC) descriptor. Extensive experimental results from 13 challenging representative texture datasets show that the proposed descriptors, applied on each dataset, can achieve interesting classification accuracy, which is competitive or better than a great number of state-of-the-art LBP variants and non-LBP methods. Furthermore, statistical hypothesis testing is performed to prove the statistical significance of the achieved accuracy improvement over all the tested datasets.

Energy Level-Based Abnormal Crowd Behavior Detection.

The change of crowd energy is a fundamental measurement for describing a crowd behavior. In this paper, we present a crowd abnormal detection method based on the change of energy-level distribution. The method can not only reduce the camera perspective effect, but also detect crowd abnormal behavior in time. Pixels in the image are treated as particles, and the optical flow method is adopted to extract the velocities of particles. The qualities of different particles are distributed as different value according to the distance between the particle and the camera to reduce the camera perspective effect. Then a crowd motion segmentation method based on flow field texture representation is utilized to extract the motion foreground, and a linear interpolation calculation is applied to pedestrian’s foreground area to determine their distance to the camera. This contributes to the calculation of the particle qualities in different locations. Finally, the crowd behavior is analyzed according to the change of the consistency, entropy and contrast of the three descriptors for co-occurrence matrix. By calculating a threshold, the timestamp when the crowd abnormal happens is determined. In this paper, multiple sets of videos from three different scenes in UMN dataset are employed in the experiment. The results show that the proposed method is effective in characterizing anomalies in videos.

An Improved Version of Texture-based Foreground Segmentation: Block-based Adaptive Segmenter

Foreground segmentation is one of moving object detection techniques of computer vision applications. To date, modern moving object detection methods require complex background modeling and thresholds tuning to confront illumination changes. This paper proposes an adaptive approach based on non-overlapping block texture representation. It aims to design a computationally light and efficient solution to improve the robustness of detection. We evaluate our proposed method on internal and public sequences and provide the quantitative and qualitative measurements. Experimental results show that the proposed method can improve the results of previous method and suitable for real-time challenges.

Fabric texture representation using the stable learned discrete cosine transform dictionary

To obtain a stable fabric texture representation result and improve the computation speed, a novel method based on dictionary learning is presented. The dictionary is learned by the alternating least-squares method using discrete cosine transform (DCT) as the initiation dictionary. To test the effectiveness of the dictionary, we comprehensively investigated 42 diverse fault-free woven fabric samples, and three fabrics with defects. After the preprocessing procedure, the woven fabric samples were characterized by the learned dictionary. The experiments on 37 samples with different fabric densities demonstrated that the Peak Signal to Noise Ratio becomes larger while the Root Mean Square Error (RMSE) diminishes as the weaving density increases. For defect fabric samples, the proposed algorithm can efficiently inspect the different types of fabric flaws. Results revealed that the learned dictionary is stable, highly efficient, and suitable for modeling fabric textures. In addition, the algorithm was validated by comparing it with the K-Singular Value Decomposition dictionary and the DCT dictionary. The learned dictionary presented strikingly better results in terms of calculation speed, consistent results, and RMSE. In general, the proposed method can be applied in studying the influence of fabric density on the representation of the fabric texture and detecting fabric flaws.

Local Concave-and-Convex Micro-Structure Patterns for texture classification

Motivated by researching new image texture modeling that improves state-of-the-art LBP variants and non-LBP descriptors, this paper proposes a novel approach for constructing local image descriptors, which are suitable for histogram based image representation. Instead of heuristic code constructions, the proposed approach is based on local concave-and-convex characteristics, which have high ability to extract discriminative and stable texture representation. Different from the majority of descriptors that only encode relationships between the pixels in doublets around central pixel (within 3 × 3 neighborhood), the proposed approach encodes relationships between the pixels in triplets by including the central pixel in the modeling. We build two distinct descriptors by dividing local features into two distinct groups, i.e., local concave and convex microstructure patterns (LCvMSP and LCxMSP), according to relationships between the pixels inside the triplets, formed along closed path around the central pixel of a 3 × 3-grayscale image patch. To make the descriptors more insensitive to noise and invariant to monotonic gray scale transformation, two supplementary triplets are added in the modeling. These triplets are formed using the central pixel and four virtual pixels set to the median of the grey-scale values of the 3 × 3 neighbourhood and the whole image and the average local and global gray levels respectively. The histograms obtained from the single scale descriptors LCvMSP and LCxMSP are concatenated together to build multi-scale histogram feature vector referred to as local concave-and-convex micro-structure pattern (LCCMSP), that is expected to better represent salient local texture structure. We evaluated the effectiveness of the proposed methods on thirteen challenging representative widely-used texture datasets, and found that the proposed LCvMSP, LCxMSP and LCCMSP operators achieve performances that are competitive or better than a large number of recent most promising state-of- the-art LBP variants and non-LBP descriptors. Statistical comparison based on Wilcoxon signed rank test demonstrated that the proposed methods are the top three over all the tested datasets.

Local binary pattern variants-based adaptive texture features analysis for posed and nonposed facial expression recognition

Facial expression recognition (FER) is an important task for various computer vision applications. The task becomes challenging when it requires the detection and encoding of macro- and micropatterns of facial expressions. We present a two-stage texture feature extraction framework based on the local binary pattern (LBP) variants and evaluate its significance in recognizing posed and nonposed facial expressions. We focus on the parametric limitations of the LBP variants and investigate their effects for optimal FER. The size of the local neighborhood is an important parameter of the LBP technique for its extraction in images. To make the LBP adaptive, we exploit the granulometric information of the facial images to find the local neighborhood size for the extraction of center-symmetric LBP (CS-LBP) features. Our two-stage texture representations consist of an LBP variant and the adaptive CS-LBP features. Among the presented two-stage texture feature extractions, the binarized statistical image features and adaptive CS-LBP features were found showing high FER rates. Evaluation of the adaptive texture features shows competitive and higher performance than the nonadaptive features and other state-of-the-art approaches, respectively.

The Effect of Contact Force on the Responses of Tactile Nerve Fibers to Scanned Textures

The perception of fine textures relies on highly precise and repeatable spiking patterns evoked in tactile afferents. These patterns have been shown to depend not only on the surface microstructure and material but also on the speed at which it moves across the skin. Interestingly, the perception of texture is independent of scanning speed, implying the existence of downstream neural mechanisms that correct for scanning speed in interpreting texture signals from the periphery. What force is applied during texture exploration also has negligible effects on how the surface is perceived, but the consequences of changes in contact force on the neural responses to texture have not been described. In the present study, we measure the signals evoked in tactile afferents of macaques to a diverse set of textures scanned across the skin at two different contact forces and find that responses are largely independent of contact force over the range tested. We conclude that the force invariance of texture perception reflects the force independence of texture representations in the nerve.