A5092447118- Advanced Neural Network Applications
- Fault Detection and Control Systems
- Advanced Memory and Neural Computing
- Photoreceptor and optogenetics research
- Conducting polymers and applications
- Building Energy and Comfort Optimization
- Time Series Analysis and Forecasting
- Energy Load and Power Forecasting
- Advanced Algorithms and Applications
- Brain Tumor Detection and Classification
- Solar Radiation and Photovoltaics
- CCD and CMOS Imaging Sensors
- Adversarial Robustness in Machine Learning
- Domain Adaptation and Few-Shot Learning
- Advanced Control Systems Optimization
- Advanced Sensor and Control Systems
- Water Quality Monitoring and Analysis
- Advanced Image and Video Retrieval Techniques
State Key Laboratory of Industrial Control Technology
2025
Northeast Normal University
2023-2024
Abstract Spiking neural network (SNN), widely known as the third-generation network, has been frequently investigated due to its excellent spatiotemporal information processing capability, high biological plausibility, and low energy consumption characteristics. Analogous working mechanism of human brain, SNN system transmits through spiking action neurons. Therefore, artificial neurons are critical building blocks for constructing in hardware. Memristors drawing growing attention...
Centrifugal compressors (CCs) are critical equipment for compressing and transporting various gases in chemical production. As a type of vane compressor, their stable reliable operation can be compromised by issues such as surge, rotating stall, other malfunctions. Traditional diagnostic approaches CCs typically depend on singular signal, vibration analysis, which one-sided overlook substantial amounts information. Additionally, multi-source-signal fusion methods relatively scarce. With this...
Objective: To analyze the risk factors of anxiety in young hypertensive patients and build a prediction model to provide scientific basis for clinical diagnosis treatment. Methods: According research content, admitted hospital from January 2022 December 2024 were selected as object at least 950 included according sample size calculation. existence anxiety, divided into control group (n = 650) observation 300), data all collected univariate analysis multivariate Logistic regression get...
FPN (Feature Pyramid Network) and transformer-based target detectors are commonly employed in detection tasks. However, these approaches suffer from design flaws that restrict their performance. To overcome limitations, we proposed TIG-DETR (Texturized Instance Guidance DETR), a novel model. comprises backbone network, TE-FPN (Texture-Enhanced FPN), an enhanced DETR detector. addresses the issue of texture information loss by utilizing bottom-up architecture, Lightweight Feature-wise...
Abstract With the increasing complexity of objects and limitation hardware resources, it is crucial to design object detection models that are both highly effective efficient. Although YOLOX, as a leading detector, strikes good balance between number parameters performance, still has some weaknesses. These include constraints in Bottleneck extracted features, insufficient feature fusion information loss FPN (Feature Pyramid Networks), an uneven trade-off performance efficiency head....