| Proj No. | A1068-251 |
| Title | Moving Object Tracking via Fusion of Underwater Optical and Sonar Images |
| Summary | Underwater target tracking plays a crucial role in various applications, including autonomous underwater vehicles (UUVs), marine exploration, underwater surveillance, and ecological monitoring. Traditional tracking methods primarily rely on either optical or sonar imaging. Optical images provide high-resolution details but are significantly affected by water turbidity, limited visibility, and light attenuation. In contrast, sonar imaging can operate effectively in turbid environments and at long ranges but suffers from lower resolution and higher noise levels. The fusion of underwater optical and sonar images aims to leverage the strengths of both modalities, enabling more robust and accurate target tracking in complex underwater environments. By integrating complementary information from these two sensor types, researchers can improve detection reliability, mitigate the limitations of individual imaging systems, and enhance real-time tracking performance. This project mainly involve these two tasks: Problem 1: Target Detection from Sonar and Optical Image Data:Develop a robust detection framework that extracts and identifies potential targets from both sonar and optical images. Given the distinct characteristics of sonar and optical imaging, tailored approaches must be employed to ensure accurate detection in each modality. Problem 2: Fusion of Detected Targets: Develop a fusion mechanism that integrates target detections from sonar and optical images, ensuring consistency across different modalities based on appearance, motion, and temporal coherence. Preliminaries: 1) Image Processing Techniques, particularly deep learning frameworks for object detection, such as YOLO and R-CNN. 2) Target Tracking and Estimation Techniques, including Kalman filtering and particle filtering. Programming language: Python is preferred. |
| Supervisor | Prof Su Rong (Loc:S1 > S1 B1B > S1 B1B 59, Ext: +65 67906042) |
| Co-Supervisor | - |
| RI Co-Supervisor | - |
| Lab | Centre for Advanced Robotics Technology Innovation (CARTIN) (Loc: S2.1-B3-01) |
| Single/Group: | Single |
| Area: | Intelligent Systems and Control Engineering |
| ISP/RI/SMP/SCP?: |