A5073179189- Industrial Vision Systems and Defect Detection
- Biometric Identification and Security
- Remote Sensing and LiDAR Applications
- Topic Modeling
- Wood and Agarwood Research
- Hand Gesture Recognition Systems
- Metabolomics and Mass Spectrometry Studies
- Gait Recognition and Analysis
- Dental Radiography and Imaging
- Infrared Thermography in Medicine
- Cell Image Analysis Techniques
- Data Quality and Management
- Natural Language Processing Techniques
- Computational Drug Discovery Methods
Zhejiang Lab
2024-2025
Taiyuan University of Technology
2024
Guilin University of Electronic Technology
2024
Traditional deep functional map frameworks are widely used for 3D shape matching; however, many methods fail to adaptively capture the relevant frequency information required estimation in complex scenarios, leading poor performance, especially under significant deformations. To address these challenges, we propose a novel unsupervised learning-based framework, Deep Frequency Awareness Functional Maps (DFAFM), specifically designed tackle diverse shape-matching problems. Our approach...
The supervision of novel psychoactive substances (NPSs) is a global problem, and the regulation NPSs was heavily relied on identifying structural matches in established databases. However, violators could circumvent legal oversight by altering side chain structure recognized existing methods cannot overcome inaccuracy lag supervision. In this study, we propose scaffold transformer-based NPS generation Screening (STNGS) framework to systematically identify evaluate potential NPSs. A...
Pulsed radio ultrawide band (IR-UWB) radar has garnered significant attention as a non-contact sensor for vital signs detection, particularly in disaster relief scenarios. However, accurately determining the number and distance of measured targets from radar, especially when occluded by objects, well extracting weak respiration heartbeat information complex echo signals, remains major challenge. This paper proposes Vital Signs Removal Method based on Matching Tracking (VSR-MP) to eliminate...
Accurate and efficient pixel-wise segmentation of wood panels is crucial for enabling machine vision technologies to optimize the sawing process. Traditional image algorithms often struggle with robustness accuracy in complex industrial environments. To address these challenges, this paper proposes an improved DeepLabV3+-based algorithm panel images. The model incorporates a lightweight MobileNetV3 backbone enhance feature extraction, reducing number parameters computational complexity while...