Coherence-Based Underwater Target Detection From Multiple Disparate Sonar Platforms

Abstract

This paper explores the use of multichannel coherence as a tool for detection of underwater targets from imagery captured from multiple disparate sonar systems. The use of multiple disparate sonars allows one to exploit a high-resolution sonar with good target definition while taking advantage of the clutter suppression abilities of a low-resolution broadband (BB) sonar coregistered over the same region to provide much better detection performance comparing to those of the single-sonar cases. In this paper, the standard Neyman-Pearson detector is extended to the dual disparate sonar case allowing target detection across two sensory channels simultaneously. A novel distributed detection system is also developed that exploits the use of multiple dual-sonar detectors for multiplatform target detection. Test results of the proposed detection methods are also presented on an underwater synthetic aperture sonar (SAS) imagery database containing data from three different imaging sonars operating at three different frequencies and resolutions. Test results illustrating the effectiveness of different coherent-based detection systems will be presented and benchmarked against those of two other detection methods in terms of probability of detection, false alarm rate, and the receiver operating characteristic (ROC) curve. Performance gains of about 23% in probability of detection were achieved over the benchmarked methods.