A5076013115- Machine Fault Diagnosis Techniques
- Image and Signal Denoising Methods
- Structural Health Monitoring Techniques
- Advanced Graph Neural Networks
- Blind Source Separation Techniques
- Complex Network Analysis Techniques
- Mathematical Analysis and Transform Methods
- Model Reduction and Neural Networks
- Generative Adversarial Networks and Image Synthesis
- Recommender Systems and Techniques
- Sparse and Compressive Sensing Techniques
- Fault Detection and Control Systems
- Functional Brain Connectivity Studies
- Advanced Numerical Analysis Techniques
- Advanced Computing and Algorithms
- Advanced Image Fusion Techniques
- Advanced Algorithms and Applications
- Advanced Electrical Measurement Techniques
- Graph Theory and Algorithms
- Advanced Sensor and Control Systems
- Advanced Image Processing Techniques
- Stock Market Forecasting Methods
- Advanced Decision-Making Techniques
- Gear and Bearing Dynamics Analysis
- Stochastic Gradient Optimization Techniques
Sun Yat-sen University
2015-2024
Key Laboratory of Guangdong Province
2019-2024
National University of Defense Technology
2024
Jingdezhen Ceramic Institute
2011-2024
Mianyang Central Hospital
2024
Guangdong Polytechnic of Science and Technology
2023
Zhongyuan University of Technology
2009-2021
Ministry of Water Resources of the People's Republic of China
2020
Institute of Soil and Water Conservation
2020
Institute of Computing Technology
2014
In this paper, a novel online signature verification technique based on discrete cosine transform (DCT) and sparse representation is proposed. We find new property of DCT, which can be used to obtain compact an using fixed number coefficients, leading simple matching procedures providing effective alternative deal with time series different lengths. The also extract energy features. Furthermore, attempt apply made, task-specific method for building overcomplete dictionaries proposed, then...
This study investigates the diagnostic potential of serum Kallikrein-related peptidase 8 (KLK8) in detecting left ventricular hypertrophy (LVH). A total 62 hypertensive patients with LVH, 60 without and healthy controls were analyzed. LVH was defined by echocardiography using LVMI thresholds (>95 g/m2 for females, >115 males). Serum KLK8 levels measured via ELISA, receiver operating characteristic (ROC) curve analysis assessed its performance. significantly higher (8.59 ± 1.59 ng/mL) than...
An approach to analyze the period of a signal based on Hilbert-Huang transform is presented in this paper. For an approximately periodic which contains plenty high frequency components, relation between its and main established. Our result that, for can be estimated accurately according main-frequency distribution. By applying technique texture analysis, novel method extract periodicity features image developed, used classification, segmentation, recognition other applications.
Lin et al. propose the iterative Toeplitz filters algorithm as an alternative for Empirical Mode Decomposition (EMD). In this algorithm, average of upper and lower envelopes is replaced by certain "moving average" obtained through a low-pass filter. Performing traditional sifting with such moving averages equivalent to iterating convolution (finite length filters). This paper studies convergence signals continuous variables, proves that limit function ideal high-pass filtering process.
The aim of this study is to investigate the morphology cortical gray matter in patients with end-stage renal disease (ESRD) and relationship between thickness kidney function.Three-dimensional high-resolution brain structural magnetic resonance imaging data were collected from 35 ESRD (28 men, 18-61 years old) 40 age- gender-matched healthy controls (HCs, 32 22-58 old). Vertex-wise analysis was then performed compare brains those HCs identify abnormalities former. Multiple biochemical...
This paper describes a novel method for edge feature detection of document images based on wavelet decomposition and reconstruction. By applying the technique, image becomes representation, i.e. is decomposed into set approximation coefficients detail coefficients. Discarding approximation, extraction implemented by means reconstruction technique. In consideration mutual frequency, overlapping will occur between details, multiresolution-edge with respect to an iterative procedure developed...
The fMRI signals are usually filtered before processing and analyzing. This process can result in the loss of information carried by higher frequency low fluctuation. ICA CCA two classical methods fMRI. finds statistically independent components observed data, however these physiologically uninterpretable without auxiliary procedures. decomposes sets data into component pairs some order, may be mixtures real noise. In order to obtain avoid filtering, we propose a mixed model based on CCA,...
Restricted Boltzmann machines (RBMs) are used to build deep-belief networks that widely thought be one of the first effective deep learning neural networks. This paper studies ability RBMs represent distributions over (0, 1in via softplus/hardplus RBM It is shown any distribution whose density depends on number 1's in their input can approximated with arbitrarily high accuracy by an size 2n + 1, which improves result a previous study reducing from n <sup...
Soil erosion is one of the most severe global environmental problems, and soil surveys are scientific basis for planning conservation ecological development. To improve sampling survey methods accurately rapidly estimate actual rates erosion, a Pan-Third Pole region was taken as an example to study methodology based on high-spatial-resolution remote sensing images. The units were designed using stratified variable probability systematic method. spatiotemporal characteristics into account,...