A5101735936- Advanced Vision and Imaging
- Face and Expression Recognition
- Radiomics and Machine Learning in Medical Imaging
- Advanced Image and Video Retrieval Techniques
- AI in cancer detection
- Advanced Neural Network Applications
- Face recognition and analysis
- Image Enhancement Techniques
- Sparse and Compressive Sensing Techniques
- Advanced Image Fusion Techniques
- Medical Imaging and Analysis
- Smart Agriculture and AI
- Biometric Identification and Security
- Advanced Computational Techniques and Applications
- Industrial Vision Systems and Defect Detection
- Advanced Image Processing Techniques
- Geographic Information Systems Studies
- Microbial Metabolic Engineering and Bioproduction
- Brain Tumor Detection and Classification
- Gaze Tracking and Assistive Technology
- Probiotics and Fermented Foods
- Image Processing Techniques and Applications
- Plant-Microbe Interactions and Immunity
- Ion-surface interactions and analysis
- Plant Pathogenic Bacteria Studies
Guangzhou Medical University
2021-2024
Guangxi University
2022-2024
Tianjin University of Technology and Education
2023
Southwest University of Science and Technology
2023
Third Affiliated Hospital of Guangzhou Medical University
2023
Zhejiang University
2022
Zunyi Medical University
2012-2019
Chinese Academy of Sciences
2017-2019
Institute of Engineering Thermophysics
2019
Peking University
2019
Farmers must accurately and promptly identify sugarcane leaf diseases with identical symptoms. RGB images have a beneficial function in disease identification. Nevertheless, complex backgrounds symptoms can significantly reduce the recognition accuracy robustness. To overcome these challenges, SLViT hybrid network is presented, which transformer encoder converted to flexible plug-in (LViT) that subsequently integrated into several locations of lightweight CNN architecture (SHDC). initially...
Lactococcus lactis IL1403 was used as an experimental strain to develop a chemically defined medium for study of the physiology and metabolic pathways lactococci. An leave-one-out technique employed determine necessity each 57 chemical components in development. A statistical design approach including three fractional factorial designs central composite optimize fermentation process with 21 variables composed 19 nutritional factors grouped from two environmental (initial pH temperature). For...
This paper aims to achieve high-accuracy 3D object detection, in which we propose a novel Spatial-Channel Attention Network (SCANet), two-stage detector that takes both LIDAR point clouds and RGB images as input generate estimates. The first stage is region proposal network (RPN) put forward new (SCA) module an Extension Spatial Upsample (ESU) module. Using the pyramid pooling structure global average pooling, SCA can not only effectively incorporate multi-scale context information, but also...
Abstract Optimization of fermentation media and processes is a difficult task due to the potential for high dimensionality nonlinearity. Here we develop evaluate variations on two novel highly efficient methods experimental optimization. The first approach based using truncated genetic algorithm with developing neural network model choose best experiments run. second uses information theory, along Bayesian regularized models, experiment selection. To these experimentally, used them new...
In this paper, a novel ultra-wideband (UWB) pulse shaping method was presented to generate IR-UWB waveforms which can suppress multiple narrow-band interferences. The proposed includes two steps. First, completely satisfy and fully utilize the FCC power spectral density mask, an "interim-pulse" generated by combining several Gaussian derivative pulses with proper combination coefficients. Then, such "interim-pulses" certain delay time, composite UWB waveform mitigate narrowband interferences...
Abstract Strabismus is a common ophthalmologic disease that affects approximately 1.19% to 5.0% of children; however if the detected early it can be treated effectively. Generally, automatic detection strabismus usually performed only by single feature, which is, with image deep features or ratio features. However, accuracy diagnosis feature unreliable. This study aims develop an intelligent model driven corneal light reflection photographs automatically detect children's strabismus. The...
Optically pumped atomic magnetometers operating in the spin exchange relaxation free (SERF) regime play increasingly important roles magnetoencephalography measurements. In this application, a multichannel arrangement is necessary. A method to construct system arrange multiple individual sensor modules, where each module designed one-beam configuration with necessary magnetic modulation. However, such an restricts spatial resolution, affects uniformity of characteristics, and increases...
Disentangling the interactions between cyanobacteria and associated bacterial community is important for understanding mechanisms that mediate formation of cyanobacterial blooms in freshwater ecosystems. Despite fact a metagenomic approach enables researchers to profile structure microbial communities with cyanobacteria, reconstructing genome sequences all members remains inefficient, due inherent enormous diversity. Here, we have established stable coculture system under high salinity,...
Possibility of synthesis gas (syngas) production from waste lubricant oil (WLO) through gasification process has been investigated. The objective this work is to develop a comprehensive model simulate WLO using Aspen Plus simulator. was based on Gibbs free energy minimization and validated against the published data. Increase in ER results CO, H2, CH4 reduction CO2 enhancement due partial oxidation H2 CO. Since standard chemical exergy lower caloric value than that CO both values are not...
Abstract In recent years, convolutional‐neural‐network based stereo matching methods have achieved significant gains compared to conventional in terms of both speed and accuracy. Current state‐of‐the‐art disparity estimation algorithms require many parameters large amounts computational resources are not suited for applications on edge devices. this paper, an end‐to‐end light‐weight network (LWNet) fast is proposed, which consists efficient backbone with multi‐scale feature fusion...
Abstract A novel nonlinear experimental design scheme to identify the optimum of a multidimensional system was computationally evaluated. The method, hybrid neural network‐genetic algorithm coupled with fuzzy clustering technique, developed accumulate former knowledge and suggest optimal operating conditions for future experimentation. Because uncertainty associated complex systems, several response surfaces various degrees complexity dimensionality were used as case studies evaluating new...
Eye localization can be viewed as one of the most essential step in eye tracking and fatigue driving detection. Differing from traditional detection method using Adaboost classifier, this paper we propose a novel based on face alignment. By exploiting 6 landmarks surrounding each eye, get accurate region, excluding glasses frames interference. Then is realized via calculating center gravity iris searching contours iris. Our experimental results demonstrate that our also achieve quite good...
The performance and robustness of fatigue detection largely decrease if the driver with glasses. To address this issue, paper proposes a practical method based on face alignment at 3000 FPS algorithm. Firstly, eye regions are localized by exploiting 6 landmarks surrounding each eye. Secondly, HOG features extracted calculated put into SVM classifier to recognize state. Finally, value PERCLOS is determine whether drowsy or not. An alarm will be generated closed for specified period time....
Breast cancer accounts for the largest number of patients among all cancers in world. Intervention treatment early breast can dramatically extend a woman's 5-year survival rate. However, lack public available mammography databases field Computer-aided Diagnosis and insufficient feature extraction ability from limit diagnostic performance cancer. In this paper, A novel classification algorithm based on Convolutional Neural Network (CNN) is proposed to improve mammography. multi-view network...
Multicenter medical research is becoming a new trend. However, interagency data privacy security protection has become major bottleneck of multicenter research. Therefore, to overcome this issue, the aim present study was apply self-developed privacy-preserving machine learning framework for researchers who can build models on from multiple sources, while providing both sensitive and learned model.Based Arya, novel computing platform developed by Healink, we constructed federated (FL) model...