A5009807039- Calcium Carbonate Crystallization and Inhibition
- Gait Recognition and Analysis
- Human Pose and Action Recognition
- Quantum Information and Cryptography
- Blind Source Separation Techniques
- Domain Adaptation and Few-Shot Learning
- Quantum Computing Algorithms and Architecture
- Minerals Flotation and Separation Techniques
- Quantum Mechanics and Applications
- Tensor decomposition and applications
- Pulmonary Hypertension Research and Treatments
- Advanced Graph Neural Networks
- Face and Expression Recognition
- Sparse and Compressive Sensing Techniques
- Cardiovascular Function and Risk Factors
- Multimodal Machine Learning Applications
- EEG and Brain-Computer Interfaces
- Hand Gesture Recognition Systems
- Adaptive optics and wavefront sensing
- Speech and Audio Processing
- Advanced Steganography and Watermarking Techniques
- Bioinformatics and Genomic Networks
- Advanced Neuroimaging Techniques and Applications
- Anomaly Detection Techniques and Applications
- Stellar, planetary, and galactic studies
University of Sheffield
2017-2024
Chevron (China)
2024
Dalian University of Technology
2007-2023
Guizhou University
2023
Insigneo
2020-2023
Baker Hughes (United States)
2015-2022
Wuhan University of Technology
2009-2022
Kennedy Krieger Institute
2020-2021
Nanjing University of Aeronautics and Astronautics
2021
Guangzhou Institute of Geochemistry
2011-2021
This paper introduces a multilinear principal component analysis (MPCA) framework for tensor object feature extraction. Objects of interest in many computer vision and pattern recognition applications, such as 2-D/3-D images video sequences are naturally described tensors or arrays. The proposed performs extraction by determining projection that captures most the original tensorial input variation. solution is iterative nature it proceeds decomposing problem to series multiple subproblems....
Common spatial pattern (CSP) is a popular algorithm for classifying electroencephalogram (EEG) signals in the context of brain-computer interfaces (BCIs). This paper presents regularization and aggregation technique CSP small-sample setting (SSS). Conventional based on sample-based covariance-matrix estimation. Hence, its performance EEG classification deteriorates if number training samples small. To address this concern, regularized (R-CSP) proposed, where estimation by two parameters to...
In this letter, we present a novel objective distortion measure for binary document images. This is based on the reciprocal of distance that straightforward to calculate. Our results show proposed matches well subjective evaluation by human visual perception.
Attention-deficit/hyperactivity disorder (ADHD) is one of the most prevalent psychiatric disorders in children and adults. While ADHD patients often display circadian abnormalities, underlying mechanisms are unclear. Here we found that zebrafish mutant for gene period1b ( per1b ) displays hyperactive, impulsive-like, attention deficit-like behaviors low levels dopamine, reminiscent human patients. We clock directly regulates dopamine-related genes monoamine oxidase dopamine β hydroxylase ,...
In motor imagery brain-computer interfaces (BCIs), the symmetric positive-definite (SPD) covariance matrices of electroencephalogram (EEG) signals carry important discriminative information. this paper, we intend to classify EEG by exploiting fact that space SPD endowed with Riemannian distance is a high-dimensional manifold. To alleviate overfitting and heavy computation problems associated conventional classification methods on manifold, propose framework for intrinsic sub-manifold...
This paper proposes an uncorrelated multilinear discriminant analysis (UMLDA) framework for the recognition of multidimensional objects, known as tensor objects. Uncorrelated features are desirable in tasks since they contain minimum redundancy and ensure independence features. The UMLDA aims to extract discriminative directly from tensorial data through solving a tensor-to-vector projection. solution consists sequential iterative processes based on alternating projection method, adaptive...
The common spatial patterns (CSP) algorithm is commonly used to extract discriminative filters for the classification of electroencephalogram (EEG) signals in context brain-computer interfaces (BCIs). However, CSP based on a sample-based covariance matrix estimation. Therefore, its performance limited when number available training samples small. In this paper, method considered such small-sample setting. We propose regularized (R-CSP) by incorporating principle generic learning. estimation...
This paper proposes an uncorrelated multilinear principal component analysis (UMPCA) algorithm for unsupervised subspace learning of tensorial data. It should be viewed as a extension the classical (PCA) framework. Through successive variance maximization, UMPCA seeks tensor-to-vector projection (TVP) that captures most variation in original input while producing features. The solution consists sequential iterative steps based on alternating method. In addition to deriving framework, this...
Multidimensional data (i.e., tensors) with missing entries are common in practice. Extracting features from incomplete tensors is an important yet challenging problem many fields such as machine learning, pattern recognition, and computer vision. Although the can be recovered by tensor completion techniques, these methods focus only on estimation instead of effective feature extraction. To best our knowledge, extraction has to well explored literature. In this paper, we therefore tackle...
Pulmonary arterial hypertension (PAH) is a progressive condition with high mortality. Quantitative cardiovascular magnetic resonance (CMR) imaging metrics in PAH target individual cardiac structures and have diagnostic prognostic utility but are challenging to acquire. The primary aim of this study was develop test tensor-based machine learning approach holistically identify features using CMR, secondarily, visualize interpret key discriminative associated PAH.Consecutive treatment naive...
Machine learning has been widely used to develop classification models for autism spectrum disorder (ASD) using neuroimaging data. Recently, studies have shifted towards large multi-site datasets boost the clinical applicability and statistical power of results. However, performance is hindered by heterogeneous nature agglomerative datasets. In this paper, we propose new methods Autism Brain Imaging Data Exchange (ABIDE) dataset. We firstly a second-order measure functional connectivity (FC)...
Graph convolutional network (GCN) is an effective neural model for graph representation learning. However, standard GCN suffers from three main limitations: (1) most real-world graphs have no regular connectivity and node degrees can range one to hundreds or thousands, (2) neighboring nodes are aggregated with fixed weights, (3) features within a feature vector considered equally important. Several extensions been proposed tackle the limitations respectively. This paper focuses on tackling...
Background Cardiac MRI measurements have diagnostic and prognostic value in the evaluation of cardiopulmonary disease. Artificial intelligence approaches to automate cardiac segmentation are emerging but require clinical testing. Purpose To develop evaluate a deep learning tool for quantitative functional studies assess its use prognosis patients suspected having pulmonary hypertension. Materials Methods A retrospective multicenter multivendor data set was used learning–based contouring...
Abstract With the development of drilling in deep water areas, chemical treatments encounter more and challenging operational conditions, including high temperature (>300°F), pressure (>5,000 psi), salinity (>100,000 mg/L) as well potential seawater breakthrough due to waterflood operations. To choose a best-fit scale treatment program, primary step is conduct laboratory testing for performance evaluation. A deepwater operation was expected near future. program requested...
In this paper, a multilinear formulation of the popular principal component analysis (PCA) is proposed, named as PCA (MPCA), where input can be not only vectors, but also matrices or higher-order tensors. It natural extension and analogous counterparts in MPCA to eigenvalues eigenvectors are defined. The proposed has wide range applications generalization PCA. As an example, applied problem gait recognition using novel representation called EigenTensorGait. A sequence divided into half...
Summary With the advance of new exploration and production technologies, oil gas has gone to deeper tighter formations than ever before. These developments have also brought challenges in scale prediction inhibition, such as prevention formation at high temperatures (150–200°C), pressures (1,000–1,500 bar), total dissolved solids (TDS) (>300,000 mg/L) commonly experienced these depths. This paper will discuss (1) temperatures, pressures, TDS; (2) an efficient method study nucleation...
This paper proposes a color-based video analytic system for quantifying limb movements in epileptic seizure monitoring. The utilizes colored pyjamas to facilitate segmentation and tracking. Thus, it is unobtrusive requires no sensor/marker attached patient's body. We employ Gaussian mixture models background/foreground modeling detect limbs through coarse-to-fine paradigm with graph-cut-based segmentation. Next, we estimate parameters domain knowledge guidance extract displacement...
Completing a matrix from small subset of its entries, i.e., completion is challenging problem arising many real-world applications, such as machine learning and computer vision. One popular approach to solve the based on low-rank decomposition/factorization. Low-rank decomposition-based methods often require prespecified rank, which difficult determine in practice. In this paper, we propose novel method with <italic xmlns:mml="http://www.w3.org/1998/Math/MathML"...
Face recognition has been employed in various security-related applications such as surveillance, mugshot identification, e-passport, and access control. Despite its recent advancements, privacy concern is one of several issues preventing wider deployment. In this paper, we address the for a self-exclusion scenario face recognition, through combining with simple biometric encryption scheme called helper data system. The combined system described detail focus on key binding procedure....