A novel method for the segmentation, two-dimensional motion estimation, and subsequent tracking of multiple objects in sector scan sonar image sequences has been developed for various applications such as remotely operated vehicles (ROV) obstacle avoidance, visual servoing, and underwater surveillance. This method uses frequency domain filtering to distinguish static and moving objects in the sonar scan image sequence. Optical flow calculations are then performed to obtain magnitude and direction motion estimates. These motion estimates are matched with the predicted object position, and are used as a basis for identifying corresponding objects in adjacent scans.
|Journal||IEE Colloquium (Digest)|
|Publication status||Published - 1996|
|Event||Proceedings of the 1996 IEE Colloquium on Autonomous Underwater Vehicles and Their Systems - Recent Developments and Future Prospects - London, UK|
Duration: 1 May 1996 → 1 May 1996