Innovative Algorithms for the Categorization of Mine-Like Objects Using Standard Sonar Return Data.
Navy SBIR FY2013.1

Sol No.: Navy SBIR FY2013.1
Topic No.: N131-037
Topic Title: Innovative Algorithms for the Categorization of Mine-Like Objects Using Standard Sonar Return Data.
Proposal No.: N131-037-0569
Firm: Arete Associates
P.O. Box 2607
Winnetka, California 91396-2607
Contact: Jason Seely
Phone: (303) 651-6756
Web Site:
Abstract: Mine countermeasures will be a critical mission of the Littoral Combat Ship. This capability will be fulfilled by existing and future advanced SONAR systems to detect surface and volume mines. The performance of these systems is degraded by the large numbers of false alarms due to surface, volume, and sea bottom clutter, which increase operator workload and reduce effective search rate. Aret� Associates proposes to develop an integrated false alarm mitigation strategy to improve the detection performance of existing military and COTS SONAR systems. Our approach builds upon over 30 years exploiting powerful data analytic tools and techniques for detection of weak targets in highly cluttered environments. Aret�'s in-house automated tools will be used to identify optimally discriminating spatial and spectral feature sets and feature-based classifiers to improve the performance of individual sensors. Additionally, Aret�-developed weak-target tracking techniques will be used to combine temporal data from multiple scans, or multiple sensors, to improve weak target detection and eliminate short-lived false alarms. The optimal combination of spatial, spectral and temporal information will provide a powerful and robust object classification algorithm that will probability of detection and reduce false alarms in SONAR MCM applications.
Benefits: This effort will improve detection performance and extend the mission space of existing MCM SONAR systems. The algorithms and techniques developed under these efforts will be directly applicable to non-military COTS SONAR systems used in search applications.