Matching Matrices Research Articles

Hypercorrelation Evolution for Video Class-Incremental Learning

Video class-incremental learning aims to recognize new actions while restricting the catastrophic forgetting of old ones, whose representative samples can only be saved in limited memory. Semantically variable subactions are susceptible to class confusion due to data imbalance. While existing methods address the problem by estimating and distilling the spatio-temporal knowledge, we further explores that the refinement of hierarchical correlations is crucial for the alignment of spatio-temporal features. To enhance the adaptability on evolved actions, we proposes a hierarchical aggregation strategy, in which hierarchical matching matrices are combined and jointly optimized to selectively store and retrieve relevant features from previous tasks. Meanwhile, a correlation refinement mechanism is presented to reinforce the bias on informative exemplars according to online hypercorrelation distribution. Experimental results demonstrate the effectiveness of the proposed method on three standard video class-incremental learning benchmarks, outperforming state-of-the-art methods. Code is available at: https://github.com/Lsen991031/HCE

Deep Feature Interaction Network for Point Cloud Registration, With Applications to Optical Measurement of Blade Profiles

Optical measurement methods for blade profiles attract lots of interest in industry. Due to the nature of the thin-walled and twisted spatial freeform surfaces of blades, the measurement accuracy would be significantly affected by the accumulated error associated with the geometric accuracy and motion stability of the developed multi-view system. To overcome these issues, this paper proposes a deep feature interaction network (DFINet) for fine registration of the multi-view data. In our network, we design a two-branch structure to integrate a global and a local feature extraction branch to encode point cloud features. Moreover, we propose a feature interaction module to strengthen information association between two point clouds during feature extraction. Next, an attention mechanism is used to fuse matching information between two matching matrices obtained from the global-based and the local-based features. Experimental results demonstrate the feasibility and good practical application prospect of this method.

Analysis of Coffee Shop Business Strategy in Bandar Lampung

MSMEs boost Indonesia's economy. MSMEs create jobs and reduce poverty in Indonesia, accounting for 56% of business investment and 97% of household work. The number of coffee shops indicates market interest in them. Coffee shops are now used for studying, meeting, and hanging out with friends. Teenagers visit Bandar Lampung City's MSME businesses. Shop and cafe owners benefit from the market segmentation of young people. Even though young people and teenagers love it, a coffee shop must innovate and grow to survive because so many businesses can copy it. This study used quantitative and descriptive qualitative methods. This method collects, analyzes, and interprets data. This chapter's matching and decision stage matrices use EFE, IFE, QSPM data. Early in strategy formulation, the input tool measures subjectivity. The strategist can better generate, prioritize, evaluate, and choose strategies in the Quantitative Strategic Planning matrix by making minor decisions about the relative importance of external and internal factors in the input matrix

Linking People across Text and Images Based on Social Relation Reasoning

As a sub-task of visual grounding, linking people across text and images aims to localize target people in images with corresponding sentences. Existing approaches tend to capture superficial features of people (e.g., dress and location) that suffer from the incompleteness information across text and images. We observe that humans are adept at exploring social relations to assist identifying people. Therefore, we propose a Social Relation Reasoning (SRR) model to address the aforementioned issues. Firstly, we design a Social Relation Extraction (SRE) module to extract social relations between people in the input sentence. Specially, the SRE module based on zero-shot learning is able to extract social relations even though they are not defined in the existing datasets. A Reasoning based Cross-modal Matching (RCM) module is further used to generate matching matrices by reasoning on the social relations and visual features. Experimental results show that the accuracy of our proposed SRR model outperforms the state-of-the-art models on the challenging datasets Who's Waldo and FL: MSRE, by more than 5\% and 7\%, respectively. Our source code is available at https://github.com/VILAN-Lab/SRR.

Self-feature-based point cloud registration method with a novel convolutional Siamese point net for optical measurement of blade profile

Currently, non-contact optical measurement technologies have shown prominent power in measuring blade profiles. One of the critical issues is how to align measured point clouds in different viewpoints into an entire profile. However, extracting efficient features for registration is difficult due to the thin-walled and twisted spatial freeform surfaces of blades. Besides, various densities and small overlaps between two adjacent point cloud data also increase the difficulty of registration. To overcome these issues, in this paper, laying on the foundation of a developed multi-axis motion system, we present a novel and general framework titled convolutional Siamese point net (CSPN) for partial-to-partial point cloud registration. Our network mainly consists of three parts: feature extraction, matching matrix computation and singular value decomposition (SVD). To make a balance between low-level and high-level semantics, we design a pyramidal architecture to learn the multiscale features of each point. In matching matrix computation, both feature space and coordinate space are utilized. Also, a novel attention mechanism is proposed to deal with the conflicts between matching matrices obtained by feature space and coordinate space. Experimental results demonstrate the feasibility and good practical application prospect of this method.

New process function-based selection and configuration methodology for Process Equipment Assemblies (PEAs) exemplified on the unit operation distillation

Process Equipment Assemblies (PEA) are pre-designed equipment sets which are selected instead of time-consuming tailor-made design. This paper introduces a new method for the selection and configuration of PEAs exemplified on the unit operation distillation. First step of the new methodology is the selection of the most suitable technology applying matching matrices. The matching matrix required for selecting a suitable distillation technology is introduced. Next a process function-based planning approach is executed using so called master block flowcharts (MBF). These technology specific flowcharts contain all possible process functions required for any potential separation task and contain the technology specific main function and technology independent process functions. Considering process boundary conditions and functions already available in the infrastructure, the MBF is reduced to the functions required for the process. Next, concrete PEAs are selected starting with the main function where the separation of the fluids takes place using the PEA operating range as the main selection criterion. A tool required to determine the operating range of a distillation column is demonstrated. Finally, the PEAs for auxiliary functions are selected from a databank.

Weighted Nuclear Norm Minimization-Based Regularization Method for Image Restoration

Regularization methods have been substantially applied in image restoration due to the ill-posedness of the image restoration problem. Different assumptions or priors on images are applied in the construction of image regularization methods. In recent years, matrix low-rank approximation has been successfully introduced in the image denoising problem and significant denoising effects have been achieved. Low-rank matrix minimization is an NP-hard problem and it is often replaced with the matrix’s weighted nuclear norm minimization (WNNM). The assumption that an image contains an extensive amount of self-similarity is the basis for the construction of the matrix low-rank approximation-based image denoising method. In this paper, we develop a model for image restoration using the sum of block matching matrices’ weighted nuclear norm to be the regularization term in the cost function. An alternating iterative algorithm is designed to solve the proposed model and the convergence analyses of the algorithm are also presented. Numerical experiments show that the proposed method can recover the images much better than the existing regularization methods in terms of both recovered quantities and visual qualities.

Tradeoff for water resources allocation based on updated probabilistic assessment of matching degree between water demand and water availability.

Water resources allocation is very important for water resources management. However, it is subject to the uncertainty in water availability (WA) or water demand (WD), as well as the pressure exerted by multi-stakeholders. Therefore, we propose a general framework as following: (i) applying Bayes theorem to develop a forecasting model for WA and WD probability distributions; (ii) constructing the matching matrix showing matching degree between WA and WD and assessing the probabilistic behavior of water resources allocation solutions based on the matching matrices; and (iii) performing the trade-off analysis among the solutions under different stakeholders' objectives to meet requirements of multi-stakeholders. Longgang River basin is selected as a case study area to demonstrate the proposed framework. Results show that, the forecast probability distributions of WA and WD may be updated timely with newly introduced data, and reflect their statistical characters well. Furthermore, the matching matrices illustrate the probabilities of the possible outcomes of each allocation solution clearly. From the probabilistic assessment; the results suggest: 21160×104 m3 diverted water are required to surely satisfy the current water demands, which is exactly the amount currently diverted for the study area. The proposed framework provides the updated probabilistic assessment for the possible outcomes, contributing to stakeholders to perform the tradeoff with each other. It makes significant contributions to address water allocation issues under uncertainty and is worthy to be applied broadly.

Homology Modeling and Characterization of Novel Stress Inducing Genes and their Split Sites Recognition in the Genome of Zebra Fish (Danio Rario) by Employing Special Consensus Pattern Matching Algorithms- using an in-Silico Method

Zebra fish has long been considered to be as a strong animal model in biology and modern genetics; however now a days its gaining lot of importance in environmental studies as well. The readily availability of entire genome sequences made to permit carrying out in silico studies at Genomic level. As everyone is known that stress is much more complex and complicated process that involves so much of gene regulations known as up regulation and down regulation, the corresponding stress proteins, broadly known as heat shock proteins. In the current study, the potential transcription factor binding sites were traced out by using bioinformatics tools and about 50 heat shock protein genes were predicted by using special alogorithms using pattern matching and position weight matrices. The 3D structure of DNA-binding domain of HSTF-1 ( Heat Shock Transcription factor-1) which is crucial for regulating heat shot proteins was traced out and builted by using homology modelling methods. The 3D structure of the heat shock transcription factor-1 and together with predicted transcription factor binding sites may be validated in future experimental works which would help us in understanding the complex responsive stress mechanisms lying in Zebra fish.

Estimating matching affinity matrices under low-rank constraints

AbstractIn this paper, we address the problem of estimating transport surplus (a.k.a. matching affinity) in high-dimensional optimal transport problems. Classical optimal transport theory specifies the matching affinity and determines the optimal joint distribution. In contrast, we study the inverse problem of estimating matching affinity based on the observation of the joint distribution, using an entropic regularization of the problem. To accommodate high dimensionality of the data, we propose a novel method that incorporates a nuclear norm regularization that effectively enforces a rank constraint on the affinity matrix. The low-rank matrix estimated in this way reveals the main factors that are relevant for matching.

On matchable subsets in abelian groups and their linear analogues

In this paper, we introduce the notions of matching matrices in groups and vector spaces, which lead to some necessary conditions for existence of acyclic matching in abelian groups and its linear analogue. We also study the linear local matching property in field extensions to find a dimension criterion for linear locally matchable bases. Moreover, we define the weakly locally matchable subspaces and we investigate their relations with matchable subspaces. We provide an upper bound for the dimension of primitive subspaces in a separable field extension. We employ MATLAB coding to investigate the existence of acyclic matchings in finite cyclic groups. Finally, a possible research problem on matchings in n-groups is presented. Our tools in this paper mix combinatorics and linear algebra.

LAM: Locality affine-invariant feature matching

False match removal is a crucial and fundamental task in photogrammetry and computer vision. This paper proposes a robust and efficient mismatch-removal algorithm based on the concepts of local barycentric coordinate (LBC) and matching coordinate matrices (MCMs), called locality affine-invariant matching (LAM). LAM is suitable for both rigid and nonrigid image matching problems. We define a novel LBC system based on area ratios, which is invariant to local affine transformations. We also present the MCMs based on the coordinates of matches, whose degeneracy is able to indicate the correctness of correspondences. Our LAM method first builds a mathematical model based on the LBCs to extract good matches that preserve local neighborhood structures. Then, LAM constructs local MCMs using the extracted reliable correspondences and identifies the correctness for the remaining matches via minimizing the rank of the MCMs. LAM has linear space and linearithmic time complexities. Extensive experiments on both rigid and nonrigid real datasets demonstrate the power of the proposed method; i.e., LAM is more robust to complex transformations compared to other methods and is two orders of magnitude faster than RANSAC under low inlier rates. The source code of the proposed LAM method will be publicly available in http://www.escience.cn/people/lijiayuan/index.html.

Intelligent Urban Positioning: Integration of Shadow Matching with 3D-Mapping-Aided GNSS Ranging

In dense urban areas, conventional Global Navigation Satellite Systems (GNSS) positioning can exhibit errors of tens of metres due to the obstruction and reflection of the signals by the surrounding buildings. By using Three-Dimensional (3D) mapping of the buildings, the accuracy can be significantly improved. This paper demonstrates the first integration of GNSS shadow matching with 3D-mapping-aided GNSS ranging. The integration is performed in the position domain, whereby separate ranging and shadow matching position solutions are computed, then combined using direction-dependent weighting. Two weighting strategies are compared, one based on the computation of ranging-based and shadow matching position error covariance matrices, and a deterministic approach based on the street azimuth. Using experimental data collected from a u-blox GNSS receiver, it is shown that both integrated position solutions are significantly more accurate than either shadow matching or 3D-mapping-aided ranging on their own. The overall Root Mean Square (RMS) horizontal accuracy obtained using covariance-based weighting was 6·1 m, a factor of four improvement on the 25·9 m obtained using conventional GNSS positioning. Results are also presented using smartphone data, where shadow matching is integrated with conventional GNSS positioning.

A new approach for improved time and position measurements for TOF-PET: Time-stamping of the photo-electrons using analogue SiPMs

Measurement of the Time-of-Flight (TOF) of the 511keV gammas brings an important reduction of statistical noise in the PET image, with higher precision time measurements producing clearer images. The common method of coupling a photodetector to scintillating crystals is to have two matching matrices, with a one-to-one coupling between the crystal and the photodetector. We propose a new geometry based on analogue strip SiPMs reading out a scintillator cut into slabs. This technique allows the time stamping of individual photo-electrons and extracts the best time resolution using a specific algorithm. Here we present the results from the first ‘slab module’ test.

Worldwide Analysis of Genetic and Linguistic Relationships of Human Populations

In this study we relate language differences on a global scale with genetic distances for the same populations. The analysis is carried out on more populations (130) but fewer genetic systems (11) than earlier studies. We constructed an overall genetic distance matrix that allowed for missing values. A separate genetic distance matrix was also computed for each genetic system, and matching matrices of linguistic and geographic distances were associated with each genetic distance matrix, because the number of populations used differed among the genetic systems studied. Significant matrix correlations between language and genetics were found for both overall genetic distances and a substantial number of genetic systems, even when the effects of geographic distances were held constant. This demonstrates a significant correspondence between genetics and language on a global scale. Genetic matrices were correlated with two different linguistic distance matrices: one with higher (supraphyletic) taxonomic structure, in which among other features sub-Saharan Africans separate from non-Africans in the basal split and the Eurasiatic superphylum is postulated; and one without such structure. The correlations yield no genetic evidence to support the proposed higher linguistic structure. UPGMA and neighbor-joining trees were constructed for linguistic and genetic data. The proposed African split pattern is not supported by these data. Both types of trees indicate a pattern of grouping of east Asians, Arctic populations, and Australian natives separating from Caucasoid and African populations.

Worldwide Analysis of Genetic and Linguistic Relationships of Human Populations

In this study we relate language differences on a global scale with genetic distances for the same populations. The analysis is carried out on more populations (130) but fewer genetic systems (11) than earlier studies. We constructed an overall genetic distance matrix that allowed for missing values. A separate genetic distance matrix was also computed for each genetic system, and matching matrices of linguistic and geographic distances were associated with each genetic distance matrix, because the number of populations used differed among the genetic systems studied. Significant matrix correlations between language and genetics were found for both overall genetic distances and a substantial number of genetic systems, even when the effects of geographic distances were held constant. This demonstrates a significant correspondence between genetics and language on a global scale. Genetic matrices were correlated with two different linguistic distance matrices: one with higher (supraphyletic) taxonomic structure, in which among other features sub-Saharan Africans separate from non-Africans in the basal split and the Eurasiatic superphylum is postulated; and one without such structure. The correlations yield no genetic evidence to support the proposed higher linguistic structure. UPGMA and neighbor-joining trees were constructed for linguistic and genetic data. The proposed African split pattern is not supported by these data. Both types of trees indicate a pattern of grouping of east Asians, Arctic populations, and Australian natives separating from Caucasoid and African populations.

Burials and Graphs

This article demonstrates the analytical potential of graph theory for understanding mortuary practices in past societies. We take advantage of social network analysis software PAJEK to model relationships among burials. The case study of the Early Bronze Age cemetery Rebesovice (Czech Republic) is used to explore the potential of the network approach to explain the contrast between the center and the periphery of the cemetery. Two hypotheses are proposed to explain this contrast: Chronological and social. The first hypothesis explains the difference between the center and the periphery as an effect of social standing, while the latter as an effect of time. The data set includes archaeological and biological data from 72 burials. We calculate simple matching distance matrices as a measure of dissimilarity among the burials based on socially and chronologically significant variables and Euclidean matrix as a measure of spatial proximity among pairs of graves. We project the results into geographic space an...

ISAR IMAGING OF NON-UNIFORM ROTATION TARGETS WITH LIMITED PULSES VIA COMPRESSED SENSING

This research introduces compressed sensing (CS) prin- ciple into inverse synthetic aperture radar (ISAR) imaging of non- uniform rotation targets, and high azimuth resolution can be achieved with limited number of pulses. Firstly, the sparsity of the echoed signal of radar targets with non-uniform rotation in certain matching Fourier domain is analyzed. Then the restricted isometry property (RIP) and incoherence of partial matching Fourier matrices are checked, following which an ISAR imaging method based on CS for both random sparse aperture and short aperture cases is proposed. In particular, consid- ering the dependence of the sparse dictionary on the relative rotation parameter, a parameter estimation method by the optimal search in fractional Fourier domain is presented. Simulation experiments verify the efiectiveness as well as superiority of the proposed imaging method over traditional methods in terms of imaging performance.

The transient response of floating elastic plates to wavemaker forcing in two dimensions

The time-dependent linear motion of a group of floating thin elastic plates is considered in a two-dimensional fluid domain. Forcing is provided by a wavemaker and the waves that transmit through the group of plates are partially reflected by a beach. Fourier transforms are used to relate the solutions in the time- and frequency-domain, where a solution is found using a combination of eigenfunction matching and transfer matrices. This allows reflections from the wavemaker and the beach to be included or excluded in a simple manner. Numerical results show that the frequency response of a single plate is significantly affected by the resonances introduced by the presence of the lateral boundaries. For a two-plate system the relative flexural response is found to be strongly dependent on plate spacing.

RELATIONSHIP BETWEEN ECOGEOGRAPHIC SAMPLING AND PHENOTYPIC DIVERSITY OF BRAZILIAN PSIDIUM GERMPLASM BASED ON CATEGORICAL DESCRIPTORS

Relationships between eco-geographic origin and phenotypic variation were established for 143 Psidium accessions, including (a general term for wild-growing Psidium spp. in Brazil) and guava (P. guajava) species, sampled in 31 different eco-geographic regions of ten Brazilian States. The accessions were characterized for 35 categorical traits according to the International Union for the Protection of New Varieties of Plants (UPOV) descriptors for P. guajava. The correlation between the cophenetic values and the simple matching matrices was 0.55, and the badness-of-fit of the multidimensional scaling was 0.30. Eighty-eight percent of the araca accessions were grouped together and they clustered according to ecogeographic regions. Clusters of guava accessions were verified only in collections from four Brazilian States, and did not present clustering by specific ecogeographic regions. It was expected that the capturing of rare alleles and the enhancement of the genetic variability of Psidium germplasm could be achieved by sampling accessions according to predefined geographical areas, even for guava germplasm.