Local Structure Patterns Research Articles

Morphometric analysis within populations of Hylomyscus walterverheyeni (Rodentia: Muridae) confirms complex relationships between small rodent populations of West Central African forests

Our recent morphometric study on the strictly forest-dwelling small rodent (Praomys misonnei) in West Central Africa revealed complex relationships between populations. The present study tests this hypothesis on another species of forest rodent (Hylomyscus walterverheyeni), and more precisely compares specimens from Kessala (Gabon) with specimens from southern Cameroon. Sixty-eight specimens were used for the morphometric analyses, and four standard external characters were measured: head + body length, tail length, hindfoot length, and ear length. Highly significant morphometric differences were observed between the H. walterverheyeni populations at the local and regional scales. A principal component analysis showed that the Gabonese H. walterverheyeni populations were morphometrically closer to one another than to the Cameroonian ones. Moreover, the body size of the Gabonese H. walterverheyeni populations was smaller than that of the Cameroonian populations. Overall, H. walterverheyeni is characterized by a local morphometric structure pattern similar to that of P. misonnei. These new results further support the scenario of complex relationships between the forest populations of West Central African rodents. This suggests the existence of persistent local biogeographical barriers despite several forest expansion episodes in the upper Pleistocene, as documented by the landscape history of West Central Africa.

The impact of oxide impurity on the structure of FLiNaK and FLiNaK–CeF3 melts: A simulation study

Oxide impurities affect the performance of the molten mixtures applicable in molten salt reactors. To reveal the underlying reasons of changes in physicochemical properties, the in-depth investigations on microscopic structure of melts are demanded. In this study, the FLiNaK and FLiNaK–CeF3 molten mixtures with small addition of Li2O were simulated within the ab initio theory to obtain the local structure patterns. The emphasis is placed on oxygen-based groupings since these are expected to be the most stable. In order to estimate the stability, the calculation of ion pair lifetimes was performed. It is shown that the most probable ion groupings existing in the FLiNaK–Li2O melt are [OLi5K], [OLi5K2] and [OLi4NaK2], while the FLiNaK–Li2O–CeF3 melt contain [OCe2Li2], [OCe2Li2K], [OCe2Li3] and [OCe3Li] units predominantly. Within the groupings, the most stable ion pairs are O-Li or O-Ce, depending on the melt composition.

On the Propensity of Excess Hydroxide Ions at the Alcohol Monolayer–Water Interface

Atomistic molecular dynamics simulations have been employed to study the self-ion (H+ and OH-) distribution at the interface between long-chain C16-OH alcohol (cetyl alcohol) monolayer and water. It is well known that the free air-water interface is acidic due to accumulation of the hydronium (H3O+) ions at the interface. In the present study, we have observed that contrary to the air-water interface, at the long-chain alcohol monolayer-water interface, it is the hydroxide (OH-) ion, not the hydronium ion (H3O+) that gets accumulated. By calculating the potential of mean forces, it is confirmed that there is extra stabilization for the OH- ions at the interface relative to the bulk, but no such stabilization is observed for the H3O+ ions. By analyzing the interaction of the self-ions with other constituents in the medium, it is clearly shown that the favorable interaction of the OH- ions with the alcoholic -OH groups stabilizes this ion at the interface. By calculating coordination numbers of the self-ions it is observed that around 50% water neighbors are substituted by alcoholic -OH in case of the hydroxide ion at the interface, whereas in the case of hydronium ions, only 15% water neighbors are substituted by the alcoholic -OH. The most interesting observation about the local structure and H-bonding pattern is that the hydroxide ion acts solely as the H-bond acceptor, but the hydronium ion acts only as the H-bond donor.

Impact of network topology on the thermal and mechanical properties of lithium germanate glasses

AbstractIn this work, we study the structure–topology–property relations of a series of melt‐quenched lithium germanate glasses. These glasses exhibit the so‐called germanate anomaly, that is, the germanium atoms feature a distribution of four‐coordinated and higher coordinated germanium species, manifesting itself as anomalies in several material properties. Here, we couple variations in the number of atomic bond constraints with measured variations in thermal and mechanical properties, including thermal conductivity, Vickers hardness, and fracture toughness. For thermal conductivity, a strong correlation is found with sound velocity as well as with the volumetric constraint density. For hardness, a good correlation with volumetric constraint density is found, whereas, for fracture toughness, variations in network topology alone are insufficient to explain the composition–property relation. To account for this, we apply a recent model which incorporates knowledge of local structure, mechanical properties, and fracture patterns to predict the fracture toughness, showing a good qualitative agreement with the experimental data.

Microbiome Data Analysis by Symmetric Non-negative Matrix Factorization With Local and Global Regularization.

A network is an efficient tool to organize complicated data. The Laplacian graph has attracted more and more attention for its good properties and has been applied to many tasks including clustering, feature selection, and so on. Recently, studies have indicated that though the Laplacian graph can capture the global information of data, it lacks the power to capture fine-grained structure inherent in network. In contrast, a Vicus matrix can make full use of local topological information from the data. Given this consideration, in this paper we simultaneously introduce Laplacian and Vicus graphs into a symmetric non-negative matrix factorization framework (LVSNMF) to seek and exploit the global and local structure patterns that inherent in the original data. Extensive experiments are conducted on three real datasets (cancer, cell populations, and microbiome data). The experimental results show the proposed LVSNMF algorithm significantly outperforms other competing algorithms, suggesting its potential in biological data analysis.

Multi-objective topology optimization for the solar thermal decomposition of methane reactor enhancement

A multi-objective topology optimization method is proposed to generate local solid structure that can enhance the solar thermal decomposition of the methane (TDM) reactor. Two dominant factors can govern the performance of solar reactor, with heat transfer entropy generation and viscous dissipation representing heat exchange capability and flow resistance, respectively. Therefore, local solid structure and flow pattern can be generated under the compromise between the extreme value of the two factors. This multi-objective topology optimization problem is solved by the calculus of variations method and globally convergent version of the method of moving asymptotes (GCMMA) with the governing equations of the solar TDM reactor as constraints. A novel local structure of the solar TDM reactor is generated, which can improve the heat exchange effect and the conversion rate of the reaction.

Discriminative Sketch Topic Model With Structural Constraint for SAR Image Classification

Synthetic aperture radar (SAR) image classification is an important part in the understanding and interpretation of SAR images. Each patch in SAR images has a scene category, but usually contains multiple land-cover classes or latent properties, which can be represented by topics in the probabilistic topic model (PTM). The representation and selection of discriminative features in PTM have a large impact on the classification results. Most of the existing feature learning methods do not make full use of high-level structure feature and the feature correlation within similar images to mine discriminative features. Therefore, this article proposes a discriminative sketch topic model with structural constraint (C-SSTM) for SAR image classification. In the proposed model, each image patch is characterized by structural and texture features. In particular, the sketch structural feature is based on the sketch map to represent the image local structure pattern. Then, the local image manifold information is preserved in terms of structure and texture. In the structural constraint, the texture and structure of each image patch are combined to learn discriminative latent semantic topics between image patches. Finally, each image patch is quantified by discriminative latent semantic topics instead of low-level representation. The experimental results tested on synthetic and real SAR images demonstrate that the proposed C-SSTM is able to learn effective structural feature representation from SAR images. Compared with other related approaches, C-SSTM produces competitive classification accuracies with high time efficiency.

Effects of climate and geography on spatial patterns of genetic structure in tropical skinks

Knowledge of how contemporary and historical factors drive patterns of genetic structure across geographic space can shed light on the processes underlying diversification. This approach is especially fruitful in studies of widespread species or species clades that occur across multiple environmental conditions and biomes. In the Neotropics, specifically, molecular data from widespread vertebrate species have revealed high levels of lineage diversity and spatial genetic structure – yet studies that explore the possible correlates of local structure patterns are lacking. We investigate the distribution of lineage diversity within two widespread South American skink species complexes of the genus Mabuya. We characterize genetic structure and diversity in these widely ranged lizards, and identify potential geographic and environmental correlates, to shed light on the processes that promote lineage diversification across the heterogeneous landscapes which they occupy. In both groups, we found mitochondrial lineages to be spatially structured along the coastal forests and the savannas of Brazil. These mtDNA patterns are, however, not shared with those inferred from nuclear DNA markers. The geographic location of major mitochondrial genetic breaks is consistent with those of other taxa, suggesting common responses to former landscape change in eastern South America, particularly along geological faults. Genetic differentiation is correlated with environmental turnover and geographic separation in one, but not in the other, group of skinks. Compared to other studies of similarly widely distributed organisms, the link between spatial environmental gradients and genetic differentiation is not as strong or consistent, suggesting a more complex history underlying current phylogeographic patterns. Our genetic data indicate the existence of yet undescribed diversity in wide-ranging lizards, and the value of phylogenetic and phylogeographic studies of similarly understudied species.

Facial Emotion Identification Based on Local Binary Pattern Feature Detector

Objectives: The emotion detection is one of the important fields in computer human interaction and this study plays significant a role for identification facial expression from the images. To identify the single emotion, need a various variability of human shapes such as pose, color, texture, expression, posture and orientation. In this study, we implement Local Binary Pattern (LBP) based filters for identifying the dynamic face textures. And moreover, this approach also provides extension and simplification. Methods/Statistical Analysis: We used built-in FER2013 datasets, the database consisting seven classes (Surprise, Fear, Angry, Neutral, Sad, Disgust, Happy). The dataset is divided into three parts testing, validation and training (15% and 70%). The Convolution neural network is trained with feature Descriptor Local Binary Pattern. Findings: The experimental results have demonstrated that local LBP representations are effective in spatial dynamic feature extraction, as they encode the information of image texture configuration while providing local structure patterns. The advantages of our approach include local processing, robustness to monotonic grayscale changes and simple computation. The results show that, the performance LBP based Convolution Neural Network (CNN) model is better than conventional CNN. This research study further helps in image classification and image processing fields. Application/Improvements: It is recommended that LBP should be used for finding the local regions or pattern from the image. The LBP computation and local processing is quite better with robustness and monotonic changes. Keywords: Convolution Neural Network (CNN), Facial Emotion, Facial Expression, Face detection, Expressions, Local Binary Pattern (LBP)

Local Ternary Cross Structure Pattern: A Color LBP Feature Extraction with Applications in CBIR

With the advent of medical endoscopes, earth observation satellites and personal phones, content-based image retrieval (CBIR) has attracted considerable attention, triggered by its wide applications, e.g., medical image analytics, remote sensing, and person re-identification. However, constructing effective feature extraction is still recognized as a challenging problem. To tackle this problem, we first propose the five-level color quantizer (FLCQ) to acquire a color quantization map (CQM). Secondly, according to the anatomical structure of the human visual system, the color quantization map (CQM) is amalgamated with a local binary pattern (LBP) map to construct a local ternary cross structure pattern (LTCSP). Third, the LTCSP is further converted into the uniform local ternary cross structure pattern (LTCSPuni) and the rotation-invariant local ternary cross structure pattern (LTCSPri) in order to cut down the computational cost and improve the robustness, respectively. Finally, through quantitative and qualitative evaluations on face, objects, landmark, textural and natural scene datasets, the experimental results illustrate that the proposed descriptors are effective, robust and practical in terms of CBIR application. In addition, the computational complexity is further evaluated to produce an in-depth analysis.

Atomic Motif Recognition in (Bio)Polymers: Benchmarks From the Protein Data Bank.

Rationalizing the structure and structure–property relations for complex materials such as polymers or biomolecules relies heavily on the identification of local atomic motifs, e.g., hydrogen bonds and secondary structure patterns, that are seen as building blocks of more complex supramolecular and mesoscopic structures. Over the past few decades, several automated procedures have been developed to identify these motifs in proteins given the atomic structure. Being based on a very precise understanding of the specific interactions, these heuristic criteria formulate the question in a way that implies the answer, by defining a list of motifs based on those that are known to be naturally occurring. This makes them less likely to identify unexpected phenomena, such as the occurrence of recurrent motifs in disordered segments of proteins, and less suitable to be applied to different polymers whose structure is not driven by hydrogen bonds, or even to polypeptides when appearing in unusual, non-biological conditions. Here we discuss how unsupervised machine learning schemes can be used to recognize patterns based exclusively on the frequency with which different motifs occur, taking high-resolution structures from the Protein Data Bank as benchmarks. We first discuss the application of a density-based motif recognition scheme in combination with traditional representations of protein structure (namely, interatomic distances and backbone dihedrals). Then, we proceed one step further toward an entirely unbiased scheme by using as input a structural representation based on the atomic density and by employing supervised classification to objectively assess the role played by the representation in determining the nature of atomic-scale patterns.

Generalized Completed Local Binary Patterns for Time-Efficient Steel Surface Defect Classification

Efficient defect classification is one of the most important preconditions to achieve online quality inspection for hot-rolled strip steels. It is extremely challenging owing to various defect appearances, large intraclass variation, ambiguous interclass distance, and unstable gray values. In this paper, a generalized completed local binary patterns (GCLBP) framework is proposed. Two variants of improved completed local binary patterns (ICLBP) and improved completed noise-invariant local-structure patterns (ICNLP) under the GCLBP framework are developed for steel surface defect classification. Different from conventional local binary patterns variants, descriptive information hidden in nonuniform patterns is innovatively excavated for the better defect representation. This paper focuses on the following aspects. First, a lightweight searching algorithm is established for exploiting the dominant nonuniform patterns (DNUPs). Second, a hybrid pattern code mapping mechanism is proposed to encode all the uniform patterns and DNUPs. Third, feature extraction is carried out under the GCLBP framework. Finally, histogram matching is efficiently accomplished by simple nearest-neighbor classifier. The classification accuracy and time efficiency are verified on a widely recognized texture database (Outex) and a real-world steel surface defect database [Northeastern University (NEU)]. The experimental results promise that the proposed method can be widely applied in online automatic optical inspection instruments for hot-rolled strip steel.

Adaptive-scale convolutional neural networks for texture image analysis

This paper proposes an effective adaptive-scale convolutional neural networks (A-SCNN) for texture image analysis. We combine the multi-scale texture image analysis with the efficient feature space of a convolutional neural network to extract characteristic texture features. These latter encode regions of adaptive sizes centered on each pixel according to different optimal scales reflecting the local structure pattern content. To fix the scale-space values accurately, the Hessian-Laplacian operator is used. Experimental results demonstrate a good performance of the proposed A-SCNN in texture classification. Particularly, the CNN based on the adaptive scale shows promising for irregular texture pattern classification, and the selective sizes of both feature maps and receptive fields can further improve the performance of the classical CNN texture discrimination ability.

Adaptive-scale convolutional neural networks for texture image analysis

This paper proposes an effective adaptive-scale convolutional neural networks (A-SCNN) for texture image analysis. We combine the multi-scale texture image analysis with the efficient feature space of a convolutional neural network to extract characteristic texture features. These latter encode regions of adaptive sizes centered on each pixel according to different optimal scales reflecting the local structure pattern content. To fix the scale-space values accurately, the Hessian-Laplacian operator is used. Experimental results demonstrate a good performance of the proposed A-SCNN in texture classification. Particularly, the CNN based on the adaptive scale shows promising for irregular texture pattern classification, and the selective sizes of both feature maps and receptive fields can further improve the performance of the classical CNN texture discrimination ability.

Low-frequency photonic band structures in graphene-like triangular metallic lattice

We study the low frequency photonic band structures in triangular metallic lattice, displaying Dirac points in the frequency spectrum, and constructed upon the lowest order regular polygonal tiles. We show that, in spite of the unfavourable geometrical conditions intrinsic to the structure symmetry, the lowest frequency photonic bands are formed by resonance modes sustained by local structure patterns, with the corresponding electric fields following a triangular distribution at low structure filling rate and a honeycomb distribution at high filling rate. For both cases, the lowest photonic bands, and thus the plasma gap, can be described in the framework of a tight binding model, and analysed in terms of local resonance modes and their mutual correlations. At high filling rate, the Dirac points and their movement following the structure deformation are described in the same framework, in relation with local structure patterns and their variations, as well as the particularity of the metallic lattice that enhances the topological anisotropy.

A variety of local structure patterns and their hybridization for accurate eye detection

In this paper, existing local structure patterns (LSPs) are reviewed and categorized into two types: intensity-based LSPs (I-LSPs) and gradient-based LSPs (G-LSPs). I-LSPs include local binary pattern (LBP), modified census transform (MCT) and generalized binary pattern (GBP) methods that compare the intensities of differently selected neighboring pixels within a 3×3 neighborhood with differently-formed reference intensities, encoding 256, 511, and 19,162 binary patterns respectively. G-LSPs include local gradient pattern (LGP), modified gradient pattern (MGP), and generalized gradient pattern (GGP) methods that compare the gradient magnitudes of differently selected neighboring pixels within a 3×3 neighborhood with differently-formed reference gradients, encoding 256, 511, and 19,162 binary patterns respectively. We extend all these LSPs to multi-scale block LSPs (MB-LSPs) that concatenate multiple block-based LSPs. Finally, we propose several hybrid LSPs that combine I-LSPs and G-LSPs by means of the AdaBoost feature selection. In experiments using AR564, BioID, ColorFERET and LFW databases to evaluate the eye detection accuracy of the proposed LSPs, the I-LSPs were good for detecting the eyes in low quality images, the G-LSPs were good for detecting the eyes in high quality images, and the hybrid LSPs achieve state-of-the-art eye detection accuracy across image qualities.

A novel scheme based on local binary pattern for dynamic texture recognition

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.

Multi-scale local structure patterns histogram for describing visual contents in social image retrieval systems

Content based image retrieval systems rely heavily on the set of features extracted from images. Effective image representation emerges as a crucial step in such systems. A key challenge in visual content representation is to reduce the so called `semantic gap'. It is the inability of existing methods to describe contents in a human-oriented way. Content representation methods inspired by the human vision system have shown promising results in image retrieval. Considerable work has been carried out during the past two decades for developing methods to extract descriptors inspired by the human vision system and attempt to retrieve visual contents efficiently according to the user needs, thereby reducing the semantic gap. Despite the extensive research being conducted in this area, limitations in current image retrieval systems still exist. This paper presents a descriptor for personalized social image collections which utilizes the local structure patterns in salient edge maps of images at multiple scales. The human visual system at the basic level is sensitive to edges, corners, intersections, and other such intensity variations in images generating local structure patterns. Analyzing these patterns at multiple scales allow the most salient fine-grained and coarse-grained features to be captured. The features are accumulated in a local structure patterns histogram to index images allowing flexible querying of visual contents. The retrieval results show that the proposed descriptor ranks well among similar state-of-the-art methods for large social image collections.

Semi-Local Structure Patterns for Robust Face Detection

In many image processing and computer vision problems, including face detection, local structure patterns such as local binary patterns (LBP) and modified census transform (MCT) have been adopted in widespread applications due to their robustness against illumination changes. However, being reliant on the local differences between neighboring pixels, they are inevitably sensitive to noise. To overcome the problem of noise-vulnerability of the conventional local structure patterns, we propose semi-local structure patterns (SLSP), a novel feature extraction method based on local region-based differences. The SLSP is robust to illumination variations, distortion, and sparse noise because it encodes the relative sizes of the central region with locally neighboring regions into a binary code. The principle of SLSP leads noise-robust expansions of LBP and MCT feature extraction frameworks. In a statistical analysis, we find that the proposed methods transform a substantial amount of random noise patterns in face images into more meaningful uniform patterns. The empirical results on the ${\rm MIT} + {\rm CMU}$ dataset and FDDB (face detection dataset and benchmark) show that the proposed semi-local patterns applied to LBP and MCT feature extraction frameworks outperform the conventional LBP and MCT features in AdaBoost-based face detectors, with much higher detection rates.

Noise-invariant structure pattern for image texture classification and retrieval

A local region in an image can be defined using centre pixel and its differences with neighboring pixels. In order to characterize different texture structure in a discriminating manner, this paper proposes twoframeworks of Noise Invariant Structure Pattern (NISP) which utilizes both the centre pixel and local and global information of an image. To replace the centre pixel, a threshold computed from adding centre pixel and intensity averages is used in the LBP code computation. For adding the magnitude information, binary patterns generated by taking thresholds involving centre pixel and local and global average contrast are adopted. Also for adding the information of individual neighborhood of a given pixel, the binary patterns generated from global thresholding of local averages are used. Based on the use of local and global information, this paper suggests two noise invariant models that are CNLP and CNGP (i.e. Completed Noise-invariant Local-structure Pattern and Global---structure Pattern). The proposed NISPs are also insensitive to noise as the centre pixel is not directly used as threshold. The proposed texture descriptors are tested on some of the representative texture databases like Outex, Curet, UIUC, Brodatz and XU ---HR. The experimental results have shown that the proposed schemes can achieve higher classification and retrieval rates while being more robust to noise.