ROBUST TRACKING OF MULTIPLE OBJECTS IN SECTOR-SCAN SONAR IMAGE SEQUENCES USING OPTICAL-FLOW MOTION ESTIMATION

The fast update rate and good performance of new generation electronic sector scanning sonars is now allowing practicable use of temporal in formation for signal processing tasks such as object classification an d motion estimation. Problems remain, however, as objects change appea rance, merge, maneuver, move in and out of the field of view, and spli t due to poor segmentation. This paper presents an approach to the seg mentation, two-dimensional motion estimation, and subsequent tracking of multiple objects in sequences of sector scan sonar images. Applicat ions such as ROV obstacle avoidance, visual servoing, and underwater s urveillance are relevant. Initially, static and moving objects are dis tinguished in the sonar image sequence using frequency-domain filterin g. Optical flow calculations are then performed on moving objects with significant size to obtain magnitude and direction motion estimates. Matches of these motion estimates, and the future positions they predi ct, are then used as a basis for identifying corresponding objects in adjacent scans. To enhance robustness, a tracking tree is constructed storing multiple possible correspondences and cumulative confidence va lues obtained from successive compatibility measures. Deferred decisio n making is then employed to enable best estimates of object tracks to be updated as subsequent scans produce new information. The method is shown to work well, with good tracking performance when objects merge , split, and change shape. The optical flow is demonstrated to give po sition prediction errors of between 10 and 50 cm (1%-5% of scan range) , with no violation of smoothness assumptions using sample rates betwe en 4 and 1 frames/s.