Similar Local Structure Research Articles

Reconstructible Nonlinear Dimensionality Reduction via Joint Dictionary Learning

This paper presents a parametric low-dimensional (LD) representation learning method that allows to reconstruct high-dimensional (HD) input vectors in an unsupervised manner. Under the assumption that the HD data and its LD representation share the same or similar local sparse structure, the proposed method achieves reconstructible dimensionality reduction via jointly learning dictionaries in both the original HD data space and its LD representation space. By regarding the sparse representation as a smooth function with respect to a specific dictionary, we construct an encoding–decoding block for learning LD representations from sparse coefficients of HD data. It is expected that this learning process preserves the desirable structure of HD data in the LD representation space, and simultaneously allows a reliable reconstruction from the LD space back to the original HD space. In addition, the proposed single layer encoding–decoding block can be easily extended to deep learning structures. Numerical experiments on both synthetic data sets and real images show that the proposed method achieves strongly competitive and robust performance in data DR, reconstruction, and synthesis, even on heavily corrupted data. The proposed method can be used as an alternative approach to compressive sensing (CS); however, it can outperform the traditional CS methods in: 1) task-driven learning problems, such as 2-D/3-D data visualization, and 2) data reconstruction at a lower dimensional space.

Urban sprawl and local fiscal burden: analysing the Spanish case

Urban sprawl is rapidly occurring in many Spanish urban areas. The objective of this paper is to evaluate how the trend of building dispersion of new residential areas may be affecting the fiscal stability of local governments in Spain. The wide diversity of the characteristics of Spanish urban areas as well as the existence of very similar local fiscal structures make this case particularly interesting. After delimiting the urban areas and the spatial unit of analysis, a precise index of urban sprawl, calculated with geo-referenced digital cartography, is used. Using the spatially disaggregated information of taxes from the Spanish National Institute for Fiscal Studies allows for a measure of fiscal burden by local areas and the ability to distinguish among types of taxes. Control variables are also available at the local level from the Spanish Census and other databases. Two methods, quantile regressions and ordinary least squares, are used in order to measure not only the average change but the heterogeneity across the distribution of the local fiscal burden associated with the changes in urban sprawl, whilst controlling for other explanatory variables in the model. The results indicate that higher levels of urban sprawl imply higher local fiscal burden. By tax categories, the phenomenon of urban sprawl particularly affects both local indirect and direct taxation. These results suggest that local decision-makers should consider urban planning as one of the fundamental tools to assure long-term local fiscal sustainability.

Discovering Proangiogenic Drugs in Ischemic Stroke Based on the Relationship between Protein Domain and Drug Substructure.

As an important protective mechanism against cerebral ischemia, angiogenesis has become a topic of interest in the treatment of ischemic stroke with the challenge that few drugs promote angiogenesis. Previous studies of the identification of drug-target interactions mainly focused on the overall structures of drugs and proteins, which limited the discovery of novel structure drugs. In this article, we proposed a new strategy for discovering proangiogenic drugs based on the assumption that drug-protein interaction is mediated by substructure and domain. First, we identified substructure-domain relationships according to the known drug-protein interactions and established the drug-substructure-domain-protein relationships of genes that are proangiogenic in brain tissue and expressed significantly during ischemic stroke. Then we quantified the intensity of interaction between each drug and each protein. Finally, we obtained 540 interactive relationships between 238 drugs and 54 genes, establishing a drug-gene network with two patterns of independent and complex drug-gene interactions. Both of the patterns facilitated finding not only drugs with the same overall structure but also drugs with a different overall structure based on the same or a similar protein spectrum. In addition, we analyzed the mechanism of action of each predicted drug and extracted drugs with similar mechanisms. In vitro, our results showed that azelnidipine, azilsartan, lercanidipine, nafcillin, and vortioxetine enhanced bEnd.3 cell proliferation, migration, tube formation, and the expression of angiogenic marker PCNA. Azelnidipine, azilsartan, lercanidipine, and nafcillin increased the level of expression of proangiogenic factor VEGF. Unlike previous studies focusing on the overall structures of drugs, our research highlighted local structural similarity, which has great potential in the search for more proangiogenic drugs with novel structures.

Exploring Protein Conformational Diversity.

The native state of proteins is composed of conformers in dynamical equilibrium. In this chapter, different issues related to conformational diversity are explored using a curated and experimentally based database called CoDNaS (Conformational Diversity in the Native State). This database is a collection of redundant structures for the same sequence. CoDNaS estimates the degree of conformational diversity using different global and local structural similarity measures. It allows the user to explore how structural differences among conformers change as a function of several structural features providing further biological information. This chapter explores the measurement of conformational diversity and its relationship with sequence divergence. Also, it discusses how proteins with high conformational diversity could affect homology modeling techniques.

Quantifying local coordination environment and structural similarity through order parameter-based site fingerprints and their application to machine learning

Quantifying local coordination environment and structural similarity through order parameter-based site fingerprints and their application to machine learning

High-quality binary fringe patterns generation by combining optimization on global and local similarity

BackgroundThe recently proposed optimized dithering techniques can improve measurement quality. However, the objective function in these optimization methods just qualifies the global similarity of the pattern while the global optimization methods ignore the influences of local structure.MethodIn order to get high quality binary dithered patterns, this paper presents an algorithm which combines optimization on global and local similarity and implements it in the whole-fringe optimization.ResultsBoth the simulation and experimental results demonstrate that the proposed algorithm can achieve binary fringe patterns with higher phase quality and less intensity difference. The fringe patterns based on the proposed method can get high quality measurement results.ConclusionThe proposed method can get high quality binary dithered patterns under different defocusing levels by combining global similarity and local structure similarity. The proposed method can improve the robustness to the amounts of defocusing.

Structural and electronic investigation of ZnO nanostructures synthesized under different environments

An explicit study of comparison on the basis of structure and electronic properties of ZnO nanostructures was discussed. ZnO synthesized by sol-gel and hydrothermal method without using any surfactant leads to the formation of two different morphologies. Rietveld crystal structure refinement of X-ray diffraction patterns confirmed the wurtzite structure of both samples. Raman spectra also confirmed the wurtzite phase formation and improved crystallinity in sample synthesized by hydrothermal route that are concurrent with results obtained from X-ray diffraction. Field-emission scanning electron microscopy revealed the formation of ZnO spherical nanoparticle structure for sol-gel method and flower like μ-structure for ZnO prepared through hydrothermal route. Williamson-Hall equations applied to study the strain and stress parameters present in the material, show the decrease in their values as the crystal size increases. Energy band gap is calculated using diffused reflectance spectroscopy. Near-edge X-ray absorption fine-structure measurements at O K- and Zn L3,2- edges simulated with FEFF9.05 code confirmed the presence of oxygen vacancies. Further extended x-ray absorption fine-structure revealed a similar local atomic structure for both samples despite having different morphologies.

Dimensionality reduction via preserving local information

Dimensionality reduction is an important technique for multimedia data in smart cities. As a significant type of multimedia data, images are being processed frequently using such technique. It is worth noting that finding neighbors is an integral part of graph embedding algorithms in dimensionality reduction. In this paper, we present an effective and efficient metric function, named local similarity preserving (LSP), which can preserve the similarity information of each sample to its homogeneous and heterogeneous neighbors. The proposed LSP function is helpful to enhance the discriminative capability of subsequent feature extractors. Based on LSP function, we also propose two novel algorithms, local similarity preserving discriminant (LSPD) algorithm, which can preserve the local similarity information and LSPD+ algorithm, which can preserve the local similarity and geometric structure information, respectively. The experimental results on digit visualization and face recognition demonstrate the effectiveness and efficiency of the proposed algorithms.

Automated segmentation of choroidal neovascularization in optical coherence tomography images using multi-scale convolutional neural networks with structure prior

Automated segmentation of choroidal neovascularization (CNV) in optical coherence tomography (OCT) images plays an important role for the treatment of CNV disease. This paper proposes multi-scale convolutional neural networks with structure prior to segment CNV from OCT data. The proposed framework consists of two stages. In the first stage, the structure prior learning method based on sparse representation-based classification and the local potential function is developed to capture the global spatial structure and local similarity structure prior. The obtained prior can be used to improve the distinctiveness between CNV and background patches. In the second stage, multi-scale CNN model with incorporation of the learned structure prior is constructed for CNV segmentation. In this stage, multi-scale analysis is used to capture effective contextual information, which is robust to varying sizes of CNV. The proposed method was evaluated on 15 spectral domain OCT data with CNV. The experimental results demonstrate the effectiveness of proposed method.

Learning Latent Factors for Community Identification and Summarization

Network communities, which are also known as network clusters, are typical latent structures in network data. Vertices in each of these communities tend to interact more and share similar features with each other. Community identification and feature summarization are significant tasks of network analytics. To perform either of the two tasks, there have been several approaches proposed, taking into the consideration of different categories of information carried by the network, e.g., edge structure, node attributes, or both aforementioned. But few of them are able to discover communities and summarize their features simultaneously. To address this challenge, we propose a novel latent factor model for community identification and summarization (LFCIS). To perform the task, the LFCIS first formulates an objective function that evaluating the overall clustering quality taking into the consideration of both edge topology and node features in the network. In the objective function, the LFCIS also adopts an effective component that ensures those vertices sharing with both similar local structures and features to be located into the same clusters. To identify the optimal cluster membership for each vertex, a convergent algorithm for updating the variables in the objective function is derived and used by LFCIS. The LFCIS has been tested with six sets of network data, including synthetic and real networks, and compared with several state-of-the-art approaches. The experimental results show that the LFCIS outperforms most of the prevalent approaches to community discovery in social networks, and the LFCIS is able to identify the latent features that may characterize those discovered communities.

Semantic Neighbor Graph Hashing for Multimodal Retrieval.

Hashing methods have been widely used for approximate nearest neighbor search in recent years due to its computational and storage effectiveness. Most existing multimodal hashing methods try to preserve the similarity relationship based on either metric distances or semantic labels in a procrustean way, while ignoring the intra-class and inter-class variations inherent in the metric space. In this paper, we propose a novel multimodal hashing method, termed as semantic neighbor graph hashing (SNGH), which aims to preserve the fine-grained similarity metric based on the semantic graph that is constructed by jointly pursuing the semantic supervision and the local neighborhood structure. Specifically, the semantic graph is constructed to capture the local similarity structure for the image modality and the text modality, respectively. Furthermore, we define a function based on the local similarity in particular to adaptively calculate multi-level similarities by encoding the intra-class and inter-class variations. After obtaining the unified hash codes, the logistic regression with kernel trick is employed to learn view-specific hash functions independently for each modality. Extensive experiments are conducted on four widely used multimodal data sets. The experimental results demonstrate the superiority of the proposed SNGH method compared with the state-of-the-art multimodal hashing methods.

Extrapolating Energetics on Clusters and Single-Crystal Surfaces to Nanoparticles by Machine-Learning Scheme

A Bayesian linear regression scheme using a local structural similarity kernel as a descriptor is used to predict the energetics of atoms and molecules on nanoparticles. Examination of the binding energies of N, O, and NO with RhAu alloy single-crystal surfaces and particles indicates that regression models predict the binding energies on nanoparticles having diameters greater than 1.5 nm within an error range of 100–150 meV when the DFT data on single-crystal surfaces are used for training. By contrast, when the DFT data on small clusters are used for training, the regression models produce an error range of 200–400 meV. Kinetic analyses using the predicted energetics of the direct decomposition of NO on RhAu nanoparticles indicate that catalytic activity increases with a decrease in the particle diameter to 2.0 nm, whereas the activity drops when the diameter decreases to 1.5 nm. Detailed examinations of the free energy diagrams and the structures of active sites indicate that the drop in catalytic acti...

Lossless medical image compression using geometry-adaptive partitioning and least square-based prediction.

To improve the compression rates for lossless compression of medical images, an efficient algorithm, based on irregular segmentation and region-based prediction, is proposed in this paper. Considering that the first step of a region-based compression algorithm is segmentation, this paper proposes a hybrid method by combining geometry-adaptive partitioning and quadtree partitioning to achieve adaptive irregular segmentation for medical images. Then, least square (LS)-based predictors are adaptively designed for each region (regular subblock or irregular subregion). The proposed adaptive algorithm not only exploits spatial correlation between pixels but it utilizes local structure similarity, resulting in efficient compression performance. Experimental results show that the average compression performance of the proposed algorithm is 10.48, 4.86, 3.58, and 0.10% better than that of JPEG 2000, CALIC, EDP, and JPEG-LS, respectively. Graphical abstract ᅟ.

Formation of Metastable Crystals from Supercooled, Supersaturated, and Supercompressed Liquids: Role of Crystal-Liquid Interfacial Free Energy

The formation mechanism of metastable crystals from metastable liquids still remains elusive, although controlling the metastability of crystals and liquids already plays a crucial role in designing new materials in physics, chemistry, biology, and materials science. This review article describes how metastable phases can be obtained by controlling temperature, concentration, and pressure. In particular, I show the role of crystal-liquid interfacial free energy in the formation of metastable crystals from metastable liquids at a given driving force. In a microscopic viewpoint, local structure similarity between the metastable crystals and liquid determines the crystal-liquid interfacial free energy, and thus the nucleation barrier for the metastable crystals. The effect of the interfacial free energy on the formation of metastable crystals from supercooled, supersaturated, and supercompressed liquids will be demonstrated with metallic liquids, aqueous solutions, and water.

Aggregating Rich Hierarchical Features for Scene Classification in Remote Sensing Imagery

Scene classification is one of the most important issues in remote sensing image processing. To obtain a high discriminative feature representation for an image to be classified, traditional methods usually consider to densely accumulate hand-crafted low-level descriptors (e.g., scale-invariant feature transform) by feature encoding techniques. However, the performance is largely limited by the hand-crafted descriptors as they are not capable of describing the rich semantic information contained in various remote sensing images. To alleviate this problem, we propose a novel method to extract discriminative image features from the rich hierarchical information contained in convolutional neural networks (CNNs). Specifically, the low-level and middle-level intermediate convolutional features are, respectively, encoded by vector of locally aggregated descriptors (VLAD) and then reduced by principal component analysis to obtain hierarchical global features; meanwhile, the fully connected features are average pooled and subsequently normalized to form new global features. The proposed encoded mixed-resolution representation (EMR) is the concatenation of all the above-mentioned global features. Due to the usage of encoding strategies (VLAD and average pooling), our method can deal with images of different sizes. In addition, to reduce the computational consumption in the training stage, we directly extract EMR from VGG-VD and ResNet pretrained on the ImageNet dataset. We show in this paper that CNNs pretrained on the natural image dataset are more easily applied to the remote sensing dataset when the local structure similarity between two datasets is higher. Experimental evaluations on the UC-Merced and Brazilian Coffee Scenes datasets demonstrate that our method is superior to the state of the art.

Learning local dictionaries and similarity structures for single image super-resolution

The central task of reconstruction-based single image super-resolution (SR) approaches is to design an effective prior to well pose the solution to unknown up-sampled image. In this paper, we present a novel single image SR method by learning a set of local dictionaries and non-local similar structures from the input low-resolution (LR) image itself. The local dictionaries are learned by segmenting structurally different regions into different clusters and then training an individual dictionary for each cluster. With the learned dictionaries and similar information, each HR pixel in the expected HR image is estimated as the weighted average of a non-local dictionary (NLD)-based regression which assembles the local structural regularity and the non-local similar redundancies. We further transform the proposed NLD-based regression model into a unified regularization term for a maximum a posteriori probability (MAP) based SR framework. Thorough experimental results carried out on five publicly available datasets indicate that the proposed SR method is promising in producing high-quality images with finer details and sharper edges in terms of both quantitative and perceptual quality assessments.

Identifying multiple influential spreaders via local structural similarity

Identifying the nodes with largest spreading influence is of significance for information diffusion, product exposure and contagious disease detection. In this letter, based on the local structural similarity, we present a method (LSS) to identify the multiple influential spreaders. Firstly, we choose the node with the largest degree as the first spreader. Then the new spreaders would be selected if they belong to the first- or second-order–neighbor node set of the spreaders and their local structural similarities with other spreaders are smaller than the threshold parameter r. Comparing with the susceptible-infected-recovered model, the experimental results for four empirical networks show that the spreading influences of spreaders selected by the local structural similarity method are larger than that of the color method, the degree, betweenness and closeness centralities. The further experimental results for the Barabàsi-Albert and random networks show that the LSS method could identify the multiple influential spreaders more accurately, which suggests that the local structural property plays a more important role than the distance for identifying multiple influential spreaders.

Face Hallucination Using Linear Models of Coupled Sparse Support.

Most face super-resolution methods assume that low- and high-resolution manifolds have similar local geometrical structure; hence, learn local models on the low-resolution manifold (e.g., sparse or locally linear embedding models), which are then applied on the high-resolution manifold. However, the low-resolution manifold is distorted by the one-to-many relationship between low- and high-resolution patches. This paper presents the Linear Model of Coupled Sparse Support (LM-CSS) method, which learns linear models based on the local geometrical structure on the high-resolution manifold rather than on the low-resolution manifold. For this, in a first step, the low-resolution patch is used to derive a globally optimal estimate of the high-resolution patch. The approximated solution is shown to be close in the Euclidean space to the ground truth, but is generally smooth and lacks the texture details needed by the state-of-the-art face recognizers. Unlike existing methods, the sparse support that best estimates the first approximated solution is found on the high-resolution manifold. The derived support is then used to extract the atoms from the coupled low- and high-resolution dictionaries that are most suitable to learn an up-scaling function for every facial region. The proposed solution was also extended to compute face super-resolution of non-frontal images. Extensive experimental results conducted on a total of 1830 facial images show that the proposed method outperforms seven face super-resolution and a state-of-the-art cross-resolution face recognition method in terms of both quality and recognition.

Construction of 2D lateral pseudoheterostructures by strain engineering

Two-dimensional (2D) lateral heterostructures host unconventional physical properties due to their controllable band-offset tuning and interfacial sensitive characteristic. The lattice mismatch results in the difficulties to construct the perfect atomic interface in such 2D lateral heterostructures, which in turn limits their desirable properties and performances in applications. In this work, strain-modulated 2D lateral pseudoheterogeneous structures are designed and realized in the single-component 2D BiOBr nanosheets by taking advantage of their strain-sensitive crystal and electronic structures. The pseudoheterogeneous interface without atomic mismatch can be feasibly modulated by local strain distribution, which exhibits similar local electronic band structure of corresponding heterostructures. Significant enhancement in charge separation at the pseudoheterostructure was demonstrated under visible light irradiation, which is given rise to the controllable electronic band alignment across the interface. The construction of the lateral pseudoheterostructure offers a feasible and promising way to build unprecedented 2D systems with exciting properties.

A New Isomeric Porous Coordination Framework Showing Single-Crystal to Single-Crystal Structural Transformation and Preferential Adsorption of 1,3-Butadiene from C4 Hydrocarbons

Reaction of Zn(NO3)2 and 4-(3,5-dimethyl-1H-pyrazol-4-yl) benzoic acid (H2mpba) in a mixed solvent of methanol and toluene gives a new supramolecular isomer for [Zn(Hmpba)2]·guest (1·g). Single crystal X-ray diffraction analysis showed that 1·g possesses a similar local coordination structure with the three known [Zn(Hmpba)2]·guest isomers with 4-fold interpenetrated dia topologies (dissimilar interpenetration directions, void ratio varies from 0% to 28%), but displays a rare 4-connected afw topology consisting of three- and seven-membered rings, and remains a large void ratio of 36% even after 4-fold interpenetration. Interestingly, 1·g can undergo obvious framework contraction of 8.5% in the single-crystal to single-crystal manner upon guest removal to give [Zn(Hmpba)2] (1′). Comparison of the single-crystal structures of 1·g and 1′ showed that the structural transformation arises mainly from ligand conformation change. Single-component N2, CO2, and C4 hydrocarbon gas adsorption measurements for 1′ show...