Robust Matrix Research Articles

Building a robust transition matrix using causal matrix for route recommendation

Route recommendation aims to recommend the best future route based on the historical visits. However, the route data contain noise due to the GPS pseudo-range errors. While there are some works on noise removal due to the errors, the previous works handle the errors in the data collection state rather than in the data pre-processing stage. In the field of business process, heuristic miner has been used to handle data noise in the event logs. This study aims to propose a method to build a transition matrix using the causal matrix based on the dependency measure in heuristic miner. The dependency measure resulted from heuristic miner is fed into a collaborative filtering approach, that is matrix factorization, to build a model. In the end, we compare the performance of the causal matrix with a traditional frequency-based approach to see the effect of the proposed method to remove the noise. The result showed that a transition matrix with a causal matrix from heuristic miner can produce a better performance than the transition matrix without the dependency measure with a causal matrix.

TDM-MIMO Automotive Radar Point-Cloud Detection Based on the 2-D Hybrid Sparse Antenna Array

Automotive radar plays an important role in the field of advanced driver assistant systems (ADASs), in the detection of unmanned aerial vehicles (UAV), and so on. Most automotive radar products can only estimate the azimuth, and however, it is necessary for automatic driving technology to acquire height information. In this article, we propose a signal processing algorithm based on the 2-D sparse multiple-input–multiple-output (MIMO) antenna array, which can detect range, velocity, azimuth, and elevation of scattering points, and this location information can form a clear point-cloud image. First, a 2-D hybrid sparse array is designed with the advantage of reducing hardware resources, such as the number of transceiver channels. Based on the sparse antenna array, a robust matrix completion recovery algorithm, combining the low-rankness of the snapshot and the sparsity of the grid map, is further proposed to extract azimuth and elevation, and alleviate the impact of grating lobes on them. Sequentially, azimuth and elevation of targets can be shown through the correction of motion-induced phase in time-division MIMO (time-division multiplexing (TDM)-MIMO) radar and by the 2-D fast Fourier transform (2DFFT) digital beamforming (DBF). Finally, many experiments validate that a point-cloud image, such as LiDAR, can be shown, which will be hopefully applied for ADASs, UAVs, and other fields.

Detecting and Tracking Small and Dense Moving Objects in Satellite Videos: A Benchmark

Satellite video cameras can provide continuous observation for a large-scale area, which is important for many remote sensing applications. However, achieving moving object detection and tracking in satellite videos remains challenging due to the insufficient appearance information of objects and lack of high-quality datasets. In this article, we first build a large-scale satellite video dataset with rich annotations for the task of moving object detection and tracking. This dataset is collected by the Jilin-1 satellite constellation and composed of 47 high-quality videos with 1 646 038 instances of interest for object detection and 3711 trajectories for object tracking. We then introduce a motion modeling baseline to improve the detection rate and reduce false alarms based on accumulative multiframe differencing and robust matrix completion. Finally, we establish the first public benchmark for moving object detection and tracking in satellite videos and extensively evaluate the performance of several representative approaches on our dataset. Comprehensive experimental analyses and insightful conclusions are also provided. The dataset is available at <uri>https://github.com/QingyongHu/VISO</uri>.

Synthesis and evaluation of chloro SPIRO and amino SPIRO supports: Novel hydrophilic cross-linked polystyrene polymers for solid phase peptide synthesis

Synthesis of a novel, beaded, hydrophobic/hydrophilic balanced and chemically robust chloro SPIRO polymer matrix consisting of styrene and 4-Chloromethyl styrene (VBC) as backbone and 3,9-Divinyl-2,4,8,10-tetraoxaspiro[5.5]undecane as cross-linking units is demonstrated here. The suspension polymerisation technique was preferred to achieve complete monomer to polymer conversion as well as to attain spherical beaded form. Two different cross-linking densities (2 and 4 mol.%) of polymer supports were synthesized and the composition of the functional bearing site VBC was fixed as 3 mol.% for both supports. The integrity of chloro SPIRO polymer formed was verified by Fourier transform infrared spectroscopy and scanning electron microscopy analysis. Chloro SPIRO resin was subjected to Gabriel Phthalimide reaction to form amino SPIRO resin. Chemical integrity of chloro SPIRO resin was tested in different standard solid phase peptide synthetic conditions and swelling capacity was studied using amino SPIRO resin in different polar solvent conditions. The synthetic capacity of novel amino SPIRO polymer matrix was evaluated by synthesising classically challenging acyl carrier protein peptide by Fmoc strategy. The purity of peptide was checked by high performance liquid chromatography and mass by matrix assisted laser desorption.

The Role of Decidual Subpopulations in Implantation, Menstruation and Miscarriage.

In each menstrual cycle, the endometrium becomes receptive to embryo implantation while preparing for tissue breakdown and repair. Both pregnancy and menstruation are dependent on spontaneous decidualization of endometrial stromal cells, a progesterone-dependent process that follows rapid, oestrogen-dependent proliferation. During the implantation window, stromal cells mount an acute stress response, which leads to the emergence of functionally distinct decidual subsets, reflecting the level of replication stress incurred during the preceding proliferative phase. Progesterone-dependent, anti-inflammatory decidual cells (DeC) form a robust matrix that accommodates the conceptus whereas pro-inflammatory, progesterone-resistant stressed and senescent decidual cells (senDeC) control tissue remodelling and breakdown. To execute these functions, each decidual subset engages innate immune cells: DeC partner with uterine natural killer (uNK) cells to eliminate senDeC, while senDeC co-opt neutrophils and macrophages to assist with tissue breakdown and repair. Thus, successful transformation of cycling endometrium into the decidua of pregnancy not only requires continuous progesterone signalling but dominance of DeC over senDeC, aided by recruitment and differentiation of circulating NK cells and bone marrow-derived decidual progenitors. We discuss how the frequency of cycles resulting in imbalanced decidual subpopulations may determine the recurrence risk of miscarriage and highlight emerging therapeutic strategies.

Handling Cellwise Outliers by Sparse Regression and Robust Covariance

We propose a data-analytic method for detecting cellwise outliers. Given a robust covariance matrix, outlying cells (entries) in a row are found by the cellFlagger technique which combines lasso regression with a stepwise application of constructed cutoff values. The penalty term of the lasso has a physical interpretation as the total distance that suspicious cells need to move in order to bring their row into the fold. For estimating a cellwise robust covariance matrix we construct a detection-imputation method which alternates between flagging outlying cells and updating the covariance matrix as in the EM algorithm. The proposed methods are illustrated by simulations and on real data about volatile organic compounds in children.

Multi-view fuzzy clustering of deep random walk and sparse low-rank embedding

Multi-view clustering aims to improve the learning performance by exploiting discriminative information from heterogeneous data sources. It has been capturing growing research attention due to its wide utilization in the areas of data mining and computer vision. However, the full use of complementarity and consistency from multi-view data is still a challenging task. Here, we propose a novel multi-view fuzzy clustering, which adopts the joint learning of deep random walk and sparse low-rank embedding. First, a deep random walk is employed to acquire a robust similarity matrix of data points and convert fuzzy membership matrix learning to adaptive graph learning. Second, the adaptive graph is restricted with sparse low-rank constraints, which ensures its strong discriminative ability and effective cluster assignments. Third, by exploiting the sparse low-rank property, the multi-view fuzzy clustering problem is formulated as optimizing a regularized graph adjacency matrix with distance metric learning and spectral norm minimization simultaneously. The distance metric learning is embedded to scatter all data points so that any two samples are projected onto a low-dimensional subspace with a large margin. Finally, an efficient algorithm is developed for the formulated problem and its convergence is also guaranteed. In order to demonstrate the superiority of the proposed method, extensive experiments are conducted by comparing state-of-the-arts on real-world databases.

Robust Matrix Factorization With Spectral Embedding.

Nonnegative matrix factorization (NMF) and spectral clustering are two of the most widely used clustering techniques. However, NMF cannot deal with the nonlinear data, and spectral clustering relies on the postprocessing. In this article, we propose a Robust Matrix factorization with Spectral embedding (RMS) approach for data clustering, which inherits the advantages of NMF and spectral clustering, while avoiding their shortcomings. In addition, to cluster the data represented by multiple views, we present the multiview version of RMS (M-RMS), and the weights of different views are self-tuned. The main contributions of this research are threefold: 1) by integrating spectral clustering and matrix factorization, the proposed methods are able to capture the nonlinear data structure and obtain the cluster indicator directly; 2) instead of using the squared Frobenius-norm, the objectives are developed with the l2,1 -norm, such that the effects of the outliers are alleviated; and 3) the proposed methods are totally parameter-free, which increases the applicability for various real-world problems. Extensive experiments on several single-view/multiview data sets demonstrate the effectiveness of our methods and verify their superior clustering performance over the state of the arts.

Preparation and characterization of Melamine-Formaldehyde (MF)-bound cyclen derivatives for efficient removal of uranium [U(VI)] ions from wastewater

Many studies have shown successful functionalization of a polymeric matrix with different chelating moieties for metal ion removal. Melamine-Formaldehyde (MF) resin is a highly porous and robust matrix. It has appropriate chemical and physical properties to serve as a basis for an adsorbent suitable for hazardous material removal e.g., uranium. In this work, we report the synthesis and characterization of new cyclen ligands AspDOTA, GlyDOTA and PropDOTA, which are derivatives of DOTA, a well-known chelating ligand for actinide ions, and the subsequent synthesis of cyclen ligands functionalized MF resins. All new compounds were characterized (1H and 13C NMR for the ligands, MAS solid state 13C NMR for the final functionalized polymer). This study demonstrates that U(VI) extraction efficiency of the new MF-cyclen resins depends on the nature of the ligand attached to the polymeric skeleton. It is clear that the MF-AspDOTA resin is the most efficient resin for U(VI) extraction in aqueous solution at pH = 5. Therefore, it is recommended to further investigate the MF-AspDOTA system for wastewater treatment.

Radar main‐lobe jamming suppression and identification based on robust whitening Blind Source Separation and Convolutional Neural Networks

The main-lobe jamming causes serious degradation of radar system performance in practical applications. In recent years, the Blind Source Separation (BSS) method has been applied to suppress the main-lobe jamming. However, the received echoes of target and jamming cannot be effectively identified while the robustness of the conventional BSS method is poor due to the influence of system noise. To tackle these problems, a radar main-lobe jamming suppression and identification method is proposed based on robust whitening BSS and Convolutional Neural Networks methods. First, the target echoes and jamming are separated by the proposed robust whitening BSS method. An iterative optimisation method is proposed to obtain the robust whitening matrix, and sufficient conditions are provided for solvability. Then, a Target echo and Jamming Identification Network (TJINet) is proposed to identify the signals after the robust whitening BSS. The proposed TJINet method uses residual blocks and branch structures to improve classification performance and utilises asymmetric convolution blocks to extract subtle features between the target echo and deceptive jamming. Simulation and experimental results show that the proposed method has better jamming suppression performance than the traditional methods under low signal-to-noise ratio (SNR) conditions, the algorithm converges faster, and also can stably identify the target echo and jamming signals.

Pitfalls and pointers: An accessible guide to marker gene amplicon sequencing in ecological applications

Abstract Next‐Generation Sequencing (NGS) is a powerful tool that has been rapidly adopted by many ecologists studying microbial communities. Despite the exciting demonstration of NGS technology as a tool for ecological research, cryptic pitfalls inherent to its use can obscure correct interpretation of NGS data. Here, we provide an accessible overview of a NGS process that uses marker gene amplicon sequences (MGAS) that will allow scientists, particularly community ecologists, to make appropriate methodological choices and understand limits on inference about community composition and diversity that can be drawn from MGAS data. We describe the MGAS pipeline, focusing specifically on cryptic sources of variation that have received less emphasis in the ecological literature, but which may substantially impact inference about microbial community diversity and composition. By simulating communities from published microbiome data, we demonstrate how these sources of variation can generate inaccurate or misleading patterns. We specifically highlight sample dilution without researcher awareness and lane‐to‐lane variability, two cryptic sources of variation arising during the MGAS pipeline. These sources of variation affect estimates of species presence and relative abundance, particularly for species with moderate to low abundances. Each of these sources of bias can lead to errors in the estimation of both absolute and relative abundance within, and turnover among, microbial communities. Awareness and understanding of what happens and, specifically, why it happens during MGAS generation is key to generating a strong dataset and building a robust community matrix. Requesting sample dilution information from the sequencing centre, including technical replicates across sequencing lanes, and understanding how sampling intensity and community taxa distribution patterns shape the measurement of community richness, evenness and diversity are critical for drawing correct ecological inferences using MGAS data.

Robust covariance estimation for distributed principal component analysis

Fan et al. (Ann Stat 47(6):3009–3031, 2019) constructed a distributed principal component analysis (PCA) algorithm to reduce the communication cost between multiple servers significantly. However, their algorithm’s guarantee is only for sub-Gaussian data. Spurred by this deficiency, this paper enhances the effectiveness of their distributed PCA algorithm by utilizing robust covariance matrix estimators of Minsker (Ann Stat 46(6A):2871–2903, 2018) and Ke et al. (Stat Sci 34(3):454–471, 2019) to tame heavy-tailed data. The theoretical results demonstrate that when the sampling distribution is symmetric innovation with the bounded fourth moment or asymmetric with the finite 6th moment, the statistical error rate of the final estimator produced by the robust algorithm is similar to that of sub-Gaussian tails. Extensive numerical trials support the theoretical analysis and indicate that our algorithm is robust to heavy-tailed data and outliers.

Statistical confidence domain data driven based fast in-flight alignment method

Initial in-flight alignment is the basis of accurate navigation for projectiles strap-down inertial navigation system (SINS). Due to complex and highly dynamic flight environment of projectiles, inertial sensors and GNSS are susceptible to interference, which causes measurement noise appear as non-Gaussian noise, resulting in low alignment accuracy and long alignment time. Thus, this paper proposes statistical confidence domain data driven based fast in-flight alignment method. Firstly, the K matrix is used as state variables to construct projectiles initial in-flight alignment filter model. The noise evaluation indexes are defined according to measurement information and estimation results to judge abnormal degree of measurement noise. Based on this, we propose an adaptive robust matrix Kalman filter (ARMKF) method. The measurement variance matrix formulas are derived based on additive noise, which provides theoretical support for parameter selection in practical applications. Simulation and test results show that alignment accuracy and alignment time of the proposed method are better than traditional methods.

Enhanced multimodal luminescence and ultrahigh stability Eu3+-doped CsPbBr3 glasses for X-ray detection and imaging

As an emerging scintillation material, metal halide perovskite ( CsPbX 3 ) has been deemed the most potentially valuable candidate in X-ray detection and medical imaging. Nevertheless, it is a continuing challenge to implement efficient radioluminescence (RL) with high radiation stability and moisture resistance. Moreover, the optimized luminescence properties and excellent uniformity of CsPbX 3 glass are also key points for obtaining perfect X-ray images. Herein, we have successfully precipitated Eu 3 + -doped CsPbBr 3 nanocrystals (NCs) with improved photoluminescence quantum yield ( ≈ 58.6 % ) because partial Eu 3 + entered the perovskite lattice in a robust borosilicate glass matrix by in situ crystallization. The small amount of Eu addition made the lattice of NCs shrink and promoted uniform distribution of CsPbBr 3 NCs in the glass, which effectively reduced the light scattering of the sample. Subsequently, multimodal RL intensity of the CsPbBr 3 / CsPbBr 3 :xEu NCs glasses (CPB-0Eu/CPB-xEu) as a function of X-ray dose rate showed a superlinear relationship to the benefit of obtaining satisfactory X-ray images. Also, the outstanding radiation stability and water resistance of CPB-xEu were confirmed due to the protection of the robust glass matrix. Finally, an X-ray imaging system using a CPB-xEu scintillator was constructed, and the spring in the opaque sample was legibly detected under the motivation of X-rays, indicating that CsPbX 3 glasses possess extensive application prospects in terms of X-ray detection and medical imaging.

A General Framework of Remote Sensing Epipolar Image Generation

Epipolar images can improve the efficiency and accuracy of dense matching by restricting the search range of correspondences from 2-D to 1-D, which play an important role in 3-D reconstruction. As most of the satellite images in archives are incidental collections, which do not have rigorous stereo properties, in this paper, we propose a general framework to generate epipolar images for both in-track and cross-track stereo images. We first investigate the theoretical epipolar constraints of single-sensor and multi-sensor images and then introduce the proposed framework in detail. Considering large elevation changes in mountain areas, the publicly available digital elevation model (DEM) is applied to reduce the initial offsets of two stereo images. The left image is projected into the image coordinate system of the right image using the rational polynomial coefficients (RPCs). By dividing the raw images into several blocks, the epipolar images of each block are parallel generated through a robust feature matching method and fundamental matrix estimation, in which way, the horizontal disparity can be drastically reduced while maintaining negligible vertical disparity for epipolar blocks. Then, stereo matching using the epipolar blocks can be easily implemented and the forward intersection method is used to generate the digital surface model (DSM). Experimental results on several in-track and cross-track images, including optical-optical, SAR-SAR, and SAR-optical pairs, demonstrate the effectiveness of the proposed framework, which not only has obvious advantages in mountain areas with large elevation changes but also can generate high-quality epipolar images for flat areas. The generated epipolar images of a ZiYuan-3 pair in Songshan are further utilized to produce a high-precision DSM.

ROBUST PARAMETER ESTIMATION FOR RANDOM EFFECT PANEL DATA MODEL IN THE PRESENCE OF HETEROSCEDASTICITY AND INFLUENTIAL OBSERVATIONS

Panel data estimators can strongly be biased and inconsistent in the presence of heteroscedasticity and anomalous observations called influential observations (IOs) in Random effect (RE) panel data model. The existing methods (LWS, WLSF, WLSDRGP) address only the problem of IO but fail to remedy the combine problem of heteroscedasticity and IOs. Therefore, in this research we develop a method that will remedy the combine problem of heteroscedasticity and IOs based on robust heteroscedasticity consistent covariance matrix (RHCCM) estimator and fast improvised influential distance (FIID) weighting method denoted by WLSFIID. The simulation and numerical evidences show that our proposed estimation method is more efficient than the existing methods by providing smallest bias, and smallest standard error of HC4 and HC5

Scalable, colloidal synthesis of SnSb nanoalloy‐decorated mesoporous 3D NiO microspheres as a sodium‐ion battery anode

Sodium-ion batteries (SIBs) have attracted significant attention for their potential to replace lithium-ion batteries, owing to the low cost and natural abundance of sodium resources. Alloy-type anode materials like Sn and Sb possess high theoretical capacities, but they suffer from several shortcomings associated with large volume expansion and electrode disintegration during cycling. Herein, we report SnSb nanoalloy-decorated mesoporous 3-dimensional (3D) NiO microspheres that are prepared using a simple, scalable, colloidal chemistry route. The in situ formed SnSb nanoalloy offers enhanced sodiation/desodiation reversibility and improved electrode kinetics through fast ionic/electronic transport. Moreover, the mesoporous structure of the NiO microspheres serves as a robust structure-reinforcing matrix that alleviates the huge volume change of the SnSb nanoalloy and prevents particle agglomeration during sodiation/desodiation. Owing to the synergistic effects of nanosizing and nanoconfinement, this unique microsphere architecture, wherein the SnSb nanoalloys are uniformly distributed within the mesoporous 3D NiO matrix, delivers a high reversible capacity of 315 mAh g−1 with significantly improved cycle and rate performances compared with those of bulk SnSb. This facile, synthetic approach and the enhanced sodium-storage performance of our composite microspheres allow for the design and realization of high-capacity alloying-type anodes for SIBs and versatile composite materials for energy storage applications.

Component selection for exponential power mixture models

ABSTRACT Exponential Power (EP) family is a much flexible distribution family including Gaussian family as a sub-family. In this article, we study component selection and estimation for EP mixture models and regressions. The assumption on zero component mean in [X. Cao, Q. Zhao, D. Meng, Y. Chen, and Z. Xu, Robust low-rank matrix factorization under general mixture noise distributions, IEEE. Trans. Image. Process. 25 (2016), pp. 4677–4690.] is relaxed. To select components and estimate parameters simultaneously, we propose a penalized likelihood method, which can shrink mixing proportions to zero to achieve components selection. Modified EM algorithms are proposed, and the consistency of estimated component number is obtained. Simulation studies show the advantages of the proposed methods on accuracies of component number selection, parameter estimation, and density estimation. Analysis of value at risk of SHIBOR and a climate change data are given as illustration.

Robust multi-feature collective non-negative matrix factorization for ECG biometrics

The field of electrocardiogram (ECG) biometrics has received considerable attention in recent years. Although some promising methods have been proposed, it is challenging to design a robust and precise method to improve the recognition performance of ECG signals with noise and sample variation. While the advantage of improved local binary pattern (LBP) for establishing identities has been widely recognized, extracting the latent semantics from multiple LBP features has attracted little attention. We propose a robust multi-feature collective non-negative matrix factorization (RMCNMF) model to handle noise and sample variation in ECG Biometrics. We extract multiple LBP histograms as feature descriptors from segmented ECG signals, and propose a multi-feature learning framework that learns unified representations in the shared latent semantic space via collective non-negative matrix factorization. To further enhance the discrimination of learned representations, we integrate label information and multiple norms in the proposed model, which not only preserves intra- and inter-subject similarities but also mitigates the influence of noise and sample variation. RMCNMF can be solved by an efficient iteration method, for which we provide a convergence analysis in detail. Extensive experiments on four ECG databases show that it performs competitively with state-of-the-art methods.

Mitral Valve Segmentation Using Robust Nonnegative Matrix Factorization.

Analyzing and understanding the movement of the mitral valve is of vital importance in cardiology, as the treatment and prevention of several serious heart diseases depend on it. Unfortunately, large amounts of noise as well as a highly varying image quality make the automatic tracking and segmentation of the mitral valve in two-dimensional echocardiographic videos challenging. In this paper, we present a fully automatic and unsupervised method for segmentation of the mitral valve in two-dimensional echocardiographic videos, independently of the echocardiographic view. We propose a bias-free variant of the robust non-negative matrix factorization (RNMF) along with a window-based localization approach, that is able to identify the mitral valve in several challenging situations. We improve the average f1-score on our dataset of 10 echocardiographic videos by 0.18 to a f1-score of 0.56.