A5052568216- Video Surveillance and Tracking Methods
- Infrared Target Detection Methodologies
- Advanced Battery Materials and Technologies
- Advanced Measurement and Detection Methods
- Advanced Battery Technologies Research
- Advanced Neural Network Applications
- Visual Attention and Saliency Detection
- Image Retrieval and Classification Techniques
- Remote-Sensing Image Classification
- Image and Object Detection Techniques
- 3D Surveying and Cultural Heritage
- Robotics and Sensor-Based Localization
- Advancements in Battery Materials
- Electric and Hybrid Vehicle Technologies
- Target Tracking and Data Fusion in Sensor Networks
- Advanced Image and Video Retrieval Techniques
- Impact of Light on Environment and Health
- Fire Detection and Safety Systems
- Gaze Tracking and Assistive Technology
- Fuzzy Systems and Optimization
- Adversarial Robustness in Machine Learning
- Inertial Sensor and Navigation
- Blockchain Technology Applications and Security
- Advanced Sensor and Control Systems
- Advanced Image Fusion Techniques
Jiangsu University of Technology
2022-2025
Nanjing University of Science and Technology
2012-2015
Motivated by the 2013 International UAV Innovation Grand Prix, we design and implement a real-time vision system for an unmanned helicopter autonomously transferring cargoes between two platforms. In competition, four are initially placed inside circles on one platform, respectively. They required to be transferred into other platform. This paper presents core algorithms of proposed ellipse detection, tracking, single-circle-based position estimation. Experiments great success our team in...
Abstract In modern industrial fields, the quality of components is crucial for system performance. The current defect detection methods are computationally demanding and require a large number parameters. A novel lightweight method proposed, named AMD-Net, which consists three innovative modules: Attention-Guided Adaptive Compressor (AGAC), Multi-kernel Grouped Convolution(MGC) Dynamic Shared Convolution Head (DSCH). AGAC utilizes adaptive weight downsampling spatial attention mechanisms to...
Multispectral object detection is a crucial technology in remote sensing image processing, particularly low-light environments. Most current methods extract features at single scale, resulting the fusion of invalid and failure to detect small objects. To address these issues, we propose multispectral network based on multilevel feature dual modulation (GMD-YOLO). Firstly, novel dual-channel CSPDarknet53 used deep from visible-infrared images. This incorporates Ghost module, which generates...
In industrial production, the quality, reliability, and precision of parts determine overall quality performance various mechanical equipment. However, existing part defect detection methods have shortcomings in terms feature extraction fusion, leading to issues missed detection. To address this challenge, manuscript proposes a algorithm for (CRD-YOLO) based on improved YOLOv5. Our first aim is increase regional features small targets improve accuracy. manuscript, we design channel- aware...
Lane detection is a vital component of intelligent driving systems, offering indispensable functionality to keep the vehicle within its designated lane, thereby reducing risk lane departure. However, complexity traffic environment, coupled with rapid movement vehicles, creates many challenges for tasks. Current methods suffer from issues such as low feature extraction capability, poor real-time detection, and inadequate robustness. Addressing these issues, this paper proposes algorithm that...
Recent advanced trackers, composed of discriminative classification and dedicated bounding box estimation, have achieved remarkable advancements in performance visual object tracking. However, existing methods cannot satisfy the demands tracking tasks complex scenes, such as occlusion, scale variations, etc. To this end, we propose a novel online multi-scale global feature modulation for robust tracking, which is developed over accurate by overlap maximization, named ATOM+. First, coordinate...
To address the problem of deviation and registration 3D radar infrared sensor, this paper presents improves a method based on state value space federated filtering unscented Kalman filter standard filter, which conduces to real time registering system IF sensors. In presented here, covariance matching criteria-based approach was employed for judgment divergent trend, while self-adaptive attenuation factor introduced correction predicted error so as inhibit phenomenon. The experiment results...
In this paper, we propose an adaptive block-fusion multiple feature tracking algorithm in a Particle Filter framework. The features of the object are linear weighted single particle filter framework by weighting their contributions using divisibility between and background region. A modified strategy named is also devised to estimate each particle. We demonstrate color gradient histogram. Finally, several experiments implemented verify proposed algorithm.
This paper constructs an uncertain mathematical model for the fixed charge transportation problem in which several kinds of items are transported from different sources to various destinations via methods. In addition, after‐sale service such as product return and exchange that help raise customer’s satisfaction is considered during trade meet reality. Meanwhile, some indeterministic factors may occur process; probability theory not best option due lack adequate historical data. light this...
Practical application of traditional battery thermal management system (BTMS) based on only one kind phase change material (PCM) is severely limited by its poor temperature adaptability since the latent heat PCM will not take effect at temperatures both below and above narrow melting range. In this work, joint use PCMs with reasonably arranged ranges introduced to ease issue realize effective dissipation in a wide ambient Systematically numerical investigations are conducted validate...
Motivated by the 2013 International UAV Innovation Grand Prix (UAVGP), we design and implement a real-time vision system for an unmanned helicopter to transfer cargoes between two moving platforms fully autonomously. The key algorithms in include ellipse detection, tracking single-circle-based pose estimation. Comprehensive experiments have verified efficiency, accuracy robustness of algorithms. developed has helped us achieving great success UAVGP. Our team was ranked first final round competition.
Image salient object detection is useful for applications like image retrieval, recognition etc. Aiming to obtain the pixel-level saliency map, most traditional models ignore object-level information and become inappropriate in complex scenes.A new method proposed this paper by incorporating notion of background prior directly into measurement.Firstly collection proto-objects obtained adapting selective search method.Secondly each proto-object, we measure its score computing distance...