A5070007513- Robotics and Sensor-Based Localization
- Maritime Navigation and Safety
- Robotic Path Planning Algorithms
- Remote Sensing and LiDAR Applications
- Advanced Neural Network Applications
- Structural Integrity and Reliability Analysis
- Hand Gesture Recognition Systems
- Image and Object Detection Techniques
- Identification and Quantification in Food
- Advanced Chemical Sensor Technologies
- Global Cancer Incidence and Screening
- Advanced Image and Video Retrieval Techniques
- Breast Lesions and Carcinomas
- Underwater Acoustics Research
- Breast Cancer Treatment Studies
Chinese Academy of Sciences
2022-2025
Institute of Automation
2022-2025
University of Chinese Academy of Sciences
2023-2024
Shandong Institute of Automation
2023
Guangdong Institute of Intelligent Manufacturing
2022
ABSTRACT In recent years, there has been an upward trend that marine vessels, important object category in monitoring, have gradually become a research focal point the field of computer vision, such as detection, tracking, and classification. Among them, vessel re‐identification (Re‐ID) emerges significant frontier topics, which not only faces dual challenge huge intra‐class small inter‐class differences, but also complex environmental interference port monitoring scenarios. To propel...
Collaborative perception is essential for networks of agents with limited sensing capabilities, enabling them to work together by exchanging information achieve a robust and comprehensive understanding their environment. However, localization inaccuracies often lead significant spatial message displacement, which undermines the effectiveness these collaborative efforts. To tackle this challenge, we introduce FeaKM, novel method that employs Feature-level Keypoints Matching effectively...
The shipping industry has experienced rapid growth in recent years, prompting a need for advanced target recognition technology based on marine radar. This paper introduces the Track Classification Model (TCM), novel approach classifying track sequences real scenarios. TCM utilizes feature extraction network multi-feature fusion, taking radar echo images and motion information of as input, to improve classification accuracy. Additionally, also presents dataset production method that...
Lidar SLAM(simultaneous localization and mapping) is very popular due to their ability accurately localize map. This article presents a 3D-LIDAR SLAM framework suitable for multiple scenarios, which include the position estimation, LIDAR motion distortion correction, mapping method based on odometry NDT (Normal Distributions Transform), route correction function loop closure. Firstly, an integrator applied filtered inertial data estimate rotation angle. Then we do each frame of points...