A5049883225- Water Quality Monitoring Technologies
- Smart Agriculture and AI
- Advanced Sensor and Control Systems
- Advanced Image and Video Retrieval Techniques
- Water Quality Monitoring and Analysis
- Modular Robots and Swarm Intelligence
- Micro and Nano Robotics
- Image Enhancement Techniques
- Industrial Vision Systems and Defect Detection
- Advanced Computational Techniques and Applications
- Energy Efficient Wireless Sensor Networks
- Advanced Neural Network Applications
- Embedded Systems and FPGA Design
- Advanced Algorithms and Applications
- Video Surveillance and Tracking Methods
- Control and Dynamics of Mobile Robots
- Mobile Ad Hoc Networks
- Multi-Criteria Decision Making
- Image Retrieval and Classification Techniques
- Video Analysis and Summarization
- Stability and Control of Uncertain Systems
- Advanced Image Processing Techniques
- Robotic Locomotion and Control
- Human Motion and Animation
- Domain Adaptation and Few-Shot Learning
China Agricultural University
2016-2025
Ministry of Agriculture and Rural Affairs
2018-2024
National Engineering Research Center for Information Technology in Agriculture
2013-2024
Changsha University
2024
Oregon College of Oriental Medicine
2024
Oregon Medical Research Center
2024
Nanyang Institute of Technology
2024
Shandong University of Finance and Economics
2006-2023
University of Shanghai for Science and Technology
2020-2023
Xi'an Jiaotong University
2022-2023
Study ObjectivesTo determine the prevalence of insomnia, its socio-demographic and clinical correlates, treatment patterns in Chinese people.
In order to improve the prediction accuracy of dissolved oxygen in aquaculture, a hybrid model based on sparse auto-encoder (SAE) and long-short-term memory network (LSTM) is proposed this paper. The hidden layer data pre-trained by SAE contains deep latent features water quality, then input it into LSTM enhance accuracy. Experimental results show that SAE-LSTM surpasses through reducing MSE respectively 23.3%, 53.6%, 39.2% steps 3, 6, 12 hours, SAE-BPNN 87.7%, 91.9%, 90.0%, proving our more...
Most recent transformer-based models show impressive performance on vision tasks, even better than Convolution Neural Networks (CNN). In this work, we present a novel, flexible, and effective model for high-quality instance segmentation. The proposed method, Segmenting Objects with TRansformers (SOTR), simplifies the segmentation pipeline, building an alternative CNN backbone appended two parallel subtasks: (1) predicting per-instance category via transformer (2) dynamically generating mask...
The shared energy storage service provided by independent operators (IESO) has a wide range of application prospects, but when faced with the interrelated and uncertain output renewable on supply side, how to size for capacity is highly challenging problem. To this end, paper firstly proposes hybrid framework, in which private power suppliers IESO jointly provide services users. generate low-dimensional scenarios that consider correlation between multiple uncertainties wind solar generation,...
To increase prediction accuracy of dissolved oxygen (DO) in aquaculture, a hybrid model based on multi-scale features using ensemble empirical mode decomposition (EEMD) is proposed. Firstly, original DO datasets are decomposed by EEMD and we get several components. Secondly, these components used to reconstruct four terms including high frequency term, intermediate low term trend term. Thirdly, according the characteristics terms, which fluctuate violently, least squares support vector...
For both pigs in commercial farms and biological experimental at breeding bases, mounting behaviour is likely to cause damage such as epidermal wounds, lameness fractures, will no doubt reduce animal welfare. The purpose of this paper develop an efficient learning algorithm that able detect the based on data characteristics visible light images. Four Göttingen minipigs were selected subjects monitored for a week by camera overlooked pen. acquired videos analysed frames containing intercepted...
Objective— Overproduction of reactive oxygen species such as hydrogen peroxide (H 2 O ) has been implicated in various cardiovascular diseases. However, mechanism(s) underlying coronary vascular dysfunction induced by H is unclear. We studied the effect on dilation arterioles to endothelium-dependent and endothelium-independent agonists. Methods Results— Porcine were isolated pressurized without flow for vitro study. All vessels developed basal tone dilated dose-dependently activators nitric...
There has been no large-scale survey of suicide-related behaviours including suicidal ideations, plans and attempts in China involving both rural urban areas using standardized assessment tools. The aim the present study was to determine lifetime prevalence behaviour its relationship with sociodemographic factors psychiatric disorders regions Beijing, China.A total 5926 subjects were randomly selected Beijing interviewed Composite International Diagnostic Interview. Basic clinical data on...
Deep learning techniques can automatically learn features from a large number of image data set. Automatic vegetable classification is the base many applications. This paper proposed high performance method for images based on deep framework. The AlexNet network model in Caffe was used to train set obtained ImageNet and divided into training test output function adopted Rectified Linear Units (ReLU) instead traditional sigmoid tanh function, which speed up network. dropout technology improve...
Polyacrylamide (PAM)-based microspheres are commonly used as water plugging and profile control agents, but the poor mechanical strength few studies on dispersion stability, both of which closely related to performance, limit application microspheres. Herein, we synthesize nanoscale PAM-based copolymer hydrogel with an inverse microemulsion copolymerization acrylamide (AM) 2-methyl-2-acrylic amide propyl sulfonic acid (AMPS) in presence vinyl-functionalized silica nanoparticles (VSNPs). The...