A5086570697- Video Surveillance and Tracking Methods
- Advanced Image and Video Retrieval Techniques
- X-ray Spectroscopy and Fluorescence Analysis
- Anomaly Detection Techniques and Applications
- Video Analysis and Summarization
- Electron and X-Ray Spectroscopy Techniques
- Text and Document Classification Technologies
- Advanced MEMS and NEMS Technologies
- Radioactivity and Radon Measurements
- Nuclear Physics and Applications
- Elevator Systems and Control
- Photonic and Optical Devices
- High voltage insulation and dielectric phenomena
- Advanced Causal Inference Techniques
- Industrial Engineering and Technologies
- Advanced Scientific Research Methods
- Fire Detection and Safety Systems
- Advanced Electron Microscopy Techniques and Applications
- Statistical Methods in Clinical Trials
- Pancreatic and Hepatic Oncology Research
- Visual Attention and Saliency Detection
- Advanced Fiber Optic Sensors
- Infrastructure Maintenance and Monitoring
- Mining and Gasification Technologies
- Heavy metals in environment
Dalian University of Technology
2023-2024
Guangzhou Education Bureau
2019
Leipzig University
2000
Shanghai Institute of Applied Physics
2000
Heavy metal pollution is a serious environmental problem facing the world, posing threat to human health and ecosystems. Activated carbon has attracted much attention due its ability adsorb heavy metals efficiently. In this paper, preparation process of activated exhaustively described, including generation biochar from biomass feedstock by physical chemical activation. Due rich pore structure surface functional groups, exhibits excellent adsorption properties. Subsequently, article...
The speed-precision tradeoff is a critical problem in visual object tracking, as it typically requires low latency and deployed on resource-constrained platforms. Existing solutions for efficient tracking primarily focus lightweight backbones or modules, which, however, come at sacrifice precision. In this article, inspired by dynamic network routing, we propose DyTrack, transformer framework tracking. Real-world scenarios exhibit varying levels of complexity. We argue that simple sufficient...
A new method combining the pattern recognition (PR) technique with micro-PIXE spectrum was used for direct assessment of lead pollution in atmosphere Shanghai City. Single aerosol particles (SAP) PM10 (<10 μm) were analyzed using nuclear microprobe. Every particle is characterized its spectrum, which can be considered fingerprint. The PR applied to trace a contaminated back source. discrimination different pollutant sources enhanced investigating individual particles. results showed that...
Existing works mainly focus on crowd and ignore the confusion regions which contain extremely similar appearance to in background, while counting needs face these two sides at same time. To address this issue, we propose a novel end-to-end trainable region discriminating erasing network called CDENet. Specifically, CDENet is composed of modules mining module (CRM) guided (GEM). CRM consists basic density estimation (BDE) network, aware bridge network. The BDE first generates primary map,...
Cross-modal metric learning is a prominent research topic that bridges the semantic heterogeneity between vision and language. Existing methods frequently utilize simple cosine or complex distance metrics to transform pairwise features into similarity score, which suffers from an inadequate inefficient capability for measurements. Consequently, we propose Generalized Structural Sparse Function dynamically capture thorough powerful relationships across modalities pair-wise while remaining...
Enrolling adequate numbers of patients into the control arm randomized controlled trials (RCTs) often presents barriers. There is interest in leveraging real‐world data (RWD) from electronic health records (EHRs) to construct external (EC) arms supplement RCT and form hybrid (HC) arms. This investigation showed use an HC second‐line metastatic pancreatic ductal adenocarcinoma (PDAC). The experimental (atezolizumab + PEGylated recombinant human hyaluronidase (Atezo PEGPH20)) was compared with...
In order to identify the sources of individual aerosol particles, we developed an identification system based on combination micro-PIXE (particle-induced X-ray emission) technique with pattern recognition (PR) technique. It uses spectrum particle as its fingerprint origin particles. A total 294 spectra collected at Shanghai City in China, were applied test PR system. These particles from several pollution and a local environment monitor site. Most identified, most probable assigned by was...
Cross-modal metric learning is a prominent research topic that bridges the semantic heterogeneity between vision and language. Existing methods frequently utilize simple cosine or complex distance metrics to transform pairwise features into similarity score, which suffers from an inadequate inefficient capability for measurements. Consequently, we propose Generalized Structural Sparse Function dynamically capture thorough powerful relationships across modalities pair-wise while remaining...
In this paper, we propose a simple yet unified single object tracking (SOT) framework, dubbed SUTrack. It consolidates five SOT tasks (RGB-based, RGB-Depth, RGB-Thermal, RGB-Event, RGB-Language Tracking) into model trained in session. Due to the distinct nature of data, current methods typically design individual architectures and train separate models for each task. This fragmentation results redundant training processes, repetitive technological innovations, limited cross-modal knowledge...
Abstract Partial discharge (PD) pattern recognition is an effective means to accurately obtain the correspondence between PD features and insulation defects. Optical detection for has strong anti-interference ability. Therefore, this paper carried out research on of optical signal. In paper, signal under four typical defects gas-insulated switchgears (GIS) are obtained by fluorescent fiber sensing system, kinds grayscale images established. The multifractal spectrum signals extracted. By...