A5100615431- Face and Expression Recognition
- Traffic control and management
- Sparse and Compressive Sensing Techniques
- Image Retrieval and Classification Techniques
- Cervical Cancer and HPV Research
- Simulation and Modeling Applications
- AI in cancer detection
- Face recognition and analysis
- Traffic Prediction and Management Techniques
- Remote-Sensing Image Classification
- Radiomics and Machine Learning in Medical Imaging
- IoT and Edge/Fog Computing
- Remote Sensing and Land Use
- Medical Imaging and Analysis
- Industrial Technology and Control Systems
- Photoacoustic and Ultrasonic Imaging
- Neural dynamics and brain function
- Advanced Algorithms and Applications
- Advanced Electrical Measurement Techniques
- Image Processing Techniques and Applications
- Nanoplatforms for cancer theranostics
- Transportation Planning and Optimization
- Non-Destructive Testing Techniques
- Vehicle License Plate Recognition
- Model Reduction and Neural Networks
Xinxiang Medical University
2025
Peking University
1993-2024
National University of Defense Technology
2023
Jilin Electric Power Research Institute (China)
2023
Beijing Institute of Graphic Communication
2021-2022
Southern Medical University
2020-2022
University of Technology Sydney
2021
Henan Provincial People's Hospital
2011-2020
Shenzhen Institutes of Advanced Technology
2020
Beijing Institute of Big Data Research
2019
In this paper we propose a novel feature selection method based on linear discriminant analysis (LDA). To view as numerical computation problem, the shows, for first time, that it is feasible to employ LDA selection. The proposed also shows different components statistically have effects result, which can be evaluated by of eigenvector. As there are multiple eigenvectors, takes small number eigenvectors into account when evaluating effect component sample data. experimental results face...
Pap smear testing has been widely used for detecting cervical cancers based on the morphological properties of cell nuclei in microscopic image. An accurate segmentation could thus improve success rate cancer screening. In this work, a method automated using Deformable Multipath Ensemble Model (D-MEM) is proposed. The approach adopts U-shaped convolutional network as backbone network, which dense blocks are to transfer feature information more effectively. To increase flexibility model, we...
This paper proposes a novel and simple multilayer feature learning method for image classification by employing the extreme machine (ELM). The proposed algorithm is composed of two stages: ELM (ML-ELM) mapping stage stage. ML-ELM recursively built alternating between map construction maximum pooling operation. In particular, input weights constructing maps are randomly generated hence need not be trained or tuned, which makes highly efficient. Moreover, operation enables to invariant certain...
Hyperspectral image classification (HSI) has been widely used in many fields. However, noise, atmospheric conditions, material distribution and other factors seriously degrade the accuracy of HSIs. To alleviate these issues, a new approach, namely adaptive weighted quaternion Zernike moments (AWQZM), is proposed, which extracts effective spatial-spectral features for pixels HSI classification. The main contributions novelties method are as follows: 1) AWQZM can adaptively set weights each...
High performance heterogeneous clusters have been widely used to commercial and scientific areas. However, huge energy consumption has prevented the further application of large-scale clusters. It is highly desirable design energy-aware scheduling algorithm for parallel applications running on In this regard, we propose a novel called Efficient-Energy based Task Clustering Scheduling (EETCS short), which can significantly conserve power by judiciously shrinking communication when allocating...
The impervious surface (IS) cover and its dynamics are key parameters in research about urban ecology. This letter proposed an automatic framework to map the IS growth end-to-end based on temporal deep learning (DL) model long time-series Landsat imagery. First, training validating datasets were auto-generated by a joint strategy. Then, DL network was designed, predicted windows. Finally, results from multi-temporal windows combined generate map. data around core of Beijing, China, is...
The calibration of the protective current transformer (CT) is particular importance, since its accuracy at high currents crucial to correct operation subsequent relay protection devices. Conventional methods have been using an electromagnetic CT which contains iron core as standard CT. big and difficult manufacture for high-current measurement, serious residual magnetism can lead excessive measurement errors. This paper proposes a system based on indirect comparison Rogowski coil, i.e....
The accuracy and stability of electronic transformers (ETs) is the guarantee safe reliable operation a power system. As an emerging technology, however, ETs’ failure rate higher than that traditional transformers. result, calibration period should be shortened. Existing methods address verification on condition ETs lines are powered off, which has long results in power-off loss. online system for voltage (EVTs) proposed this paper. Employing specially designed standard measuring unit, can...
A color image segmentation algorithm which integrates watershed with automatic seeded region growing and merging is proposed in the paper. Firstly, was transformed from RGB space to HSI space. Next, applied obtain initial effect. And then, Based on results of segmentation, some regions were selected as automatically for seeds growth by making use differences relative Euclidean distance. Finally, a combining executed avoid excessive segmentation. The method combines advantages approach,...
Unhindered transportation of substances in the brain extracellular space (ECS) is essential for maintaining function. Regulation a novel strategy treating ECS blockage-related diseases, but few techniques have been developed to date. In this study, we established approach accelerating drainage interstitial fluid (ISF) using minimally invasive surgery, which branch external carotid artery separated and implanted epidurally (i.e., epidural arterial implantation [EAI]) promote pulsation effect...
The study is to analyse the clinical data of two groups patients with different fertilisation outcomes after short-time insemination and probe into indications. A retrospective including 558 cycles in our centre during January 2009 June 2010 which were divided by method fertilisation, 472 IVF 86 Re-ICSI group according conditions compare differences between indexes average female age, male duration infertility, forward-moving sperm counts, abnormal rate, proportion primary etc. have...
The brain extracellular space (ECS), an irregular, extremely tortuous nanoscale located between cells or and blood vessels, is crucial for nerve cell survival. It plays a pivotal role in high-level functions such as memory, emotion, sensation. However, the specific form of molecular transport within ECS remain elusive. To address this challenge, paper proposes novel approach to quantitatively analyze by solving inverse problem derived from advection-diffusion equation (ADE) using...