A5100341620- Video Surveillance and Tracking Methods
- Heat Transfer and Boiling Studies
- Advanced Neural Network Applications
- Fluid Dynamics and Heat Transfer
- Heat Transfer and Optimization
- Human Pose and Action Recognition
- Robotic Path Planning Algorithms
- Fluid Dynamics and Thin Films
- Topic Modeling
- Face recognition and analysis
- Anomaly Detection Techniques and Applications
- Domain Adaptation and Few-Shot Learning
- Multimodal Machine Learning Applications
- Advanced Vision and Imaging
- Advanced Image and Video Retrieval Techniques
- Image Enhancement Techniques
- Maritime Navigation and Safety
- Fluid Dynamics and Mixing
- Radiomics and Machine Learning in Medical Imaging
- Generative Adversarial Networks and Image Synthesis
- Geothermal Energy Systems and Applications
- AI in cancer detection
- Chaos-based Image/Signal Encryption
- Optimization and Search Problems
- Microfluidic and Capillary Electrophoresis Applications
National University of Singapore
2025
Gdansk University of Physical Education and Sport
2025
Guizhou University
2025
Tianjin University
2005-2024
Suzhou University of Science and Technology
2024
Beijing Technology and Business University
2022-2024
Anhui Xinhua University
2024
Beijing Normal University
2024
Foshan University
2024
Beijing Information Science & Technology University
2023-2024
Existing shadow detection methods suffer from an intrinsic limitation in relying on limited labeled datasets, and they may produce poor results some complicated situations. To boost the performance, this paper presents a multi-task mean teacher model for semi-supervised by leveraging unlabeled data exploring learning of multiple information shadows simultaneously. be specific, we first build baseline to simultaneously detect regions, edges, count their complementary assign student network....
In this paper, we will introduce our object detection, localization and tracking system for smart mobility applications like traffic road railway environment. Firstly, an detection approach was firstly carried out within two deep learning approaches: You Only Look Once (YOLO) V3 Single Shot Detector (SSD). A comparison between the methods allows us to identify their applicability in Both performances environments were evaluated. Secondly, distance estimation based on Monodepth algorithm...
Existing segmentation methods for brain MRI data usually leverage 3D CNNs on volumes or employ 2D image slices. We discovered that while volume-based approaches well respect spatial relationships across slices, slice-based typically excel at capturing fine local features. Furthermore, there is a wealth of complementary information between their predictions. Inspired by this observation, we develop an Uncertainty-aware Multi-dimensional Mutual learning framework to learn different dimensional...
Shadow removal in a single image has received increasing attention recent years. However, removing shadows over dynamic scenes remains largely under-explored. In this paper, we propose the first data-driven video shadow model, termed PSTNet, by exploiting three essential characteristics of shadows, i.e., physical property, spatio relation, and temporal coherence. Specifically, dedicated branch was established to conduct local illumination estimation, which is more applicable for with complex...
We propose a double image encryption scheme combining DNA, parallel compressed sensing, enhanced Zigzag confusion, chaotic sparse basis matrix, and permutation. Firstly, use the logistic map to scramble Discrete Wavelet Transformation (DWT) matrix obtain DWT matrix. Next, 2D-LICM generates measurement Parallel sensing is then performed separately on two images. Then, confusion DNA are new using sequences an encrypted image. The system's initial value generated SHA-256, which has strong...
Spatial regularization (SR) is known as an effective tool to alleviate the boundary effect of correlation filter (CF), a successful visual object tracking scheme, from which number state-of-the-art trackers can be stemmed. Nevertheless, SR highly increases optimization complexity CF and its target-driven nature makes spatially-regularized may easily lose occluded targets or surrounded by other similar objects. In this paper, we propose selective spatial (SSR) for CF-tracking scheme. It...
To solve the problem of traversal multi-target path planning for an unmanned cruise ship in unknown obstacle environment lakes, this study proposed a hybrid algorithm. The algorithm can be divided into two parts. First, was transformed traveling salesman problem, and improved Grey Wolf Optimization (GWO) used to calculate sequence. GWO optimized convergence factor by introducing Beta function, which improve speed traditional Second, based on planned target sequence, D* Lite implement between...
Objective Acute exercise has the potential to influence emotional memory and cortical hemodynamics, but specific effects depend on type of exercise. This study aimed determine whether acute Tai Chi practice enhances prefrontal cortex activation compared cycling a control condition. Methods Using within-subjects crossover design, 36 healthy university students completed three interventions: Chi, cycling, resting Emotional performance was assessed before after each intervention, hemodynamics...
Bacterial communities are highly susceptible to fluctuations in amino acid content. To investigate the response of microbial daqu environmental perturbations, we employed high-throughput sequencing and statistical analyses. Samples were collected from two workshops (A B) at distinct stages fermentation storage. Our analysis, using β-nearest taxon index (βNTI), revealed that fungal community assembly is shaped by both stochastic deterministic processes. In contrast, bacterial exhibited a...
Shadow detection in a single image has received significant research interests recent years. However, much fewer works have been explored shadow over dynamic scenes. The bottleneck is the lack of well-established dataset with high-quality annotations for video detection. In this work, we collect new (ViSha), which contains 120 videos 11,685 frames, covering 60 object categories, varying lengths, and different motion/lighting conditions. All frames are annotated pixel-level mask. To best our...
Flexible sensors have attracted great attention because of their important applications in many areas. It is to monitor the surface starch-based food during processing it can provide key information related appearance, texture level, and chewiness food. However, there no report on real-time monitoring steamed bread literature. Here, we a self-adhesive stretchable compliant sensor that be mounted provides signals for steaming process. The consisting biocompatible...
Abstract Using powder‐based ink appears to be the most suitable candidate for commercializing membrane electrode assembly (MEA), while research on NPM catalyst anion exchange water electrolyzer (AEMWE) is currently insufficient, especially at high current density. Herein, a sulfur source (NiFe Layered double hydroxide adsorbed ) confinement strategy developed integrate Ni 3 S 2 onto surface of amorphous/crystalline NiFe alloy nanoparticles (denoted NiFe/Ni‐S), achieving advanced control over...