Automated Visual Detection of Small Contacts on the Horizon
Navy SBIR FY2015.1


Sol No.: Navy SBIR FY2015.1
Topic No.: N151-040
Topic Title: Automated Visual Detection of Small Contacts on the Horizon
Proposal No.: N151-040-0563
Firm: Toyon Research Corp.
6800 Cortona Drive
Goleta, California 93117-3021
Contact: Brian Stieber
Phone: (805) 968-6787
Abstract: Detection of possible hostile threats at long ranges is critical to the situational awareness of submarines. From periscope imagery, threat detection amounts to visual detection of small targets at or near the horizon. Toyon Research Corporation proposes the development of algorithms and software for automated detection of horizon contacts with minimal false alarms. Toyon will extend existing clutter suppression and track-before-detect algorithms with a novel adaptive horizon contact filter using operator feedback in an active learning framework. The proposed system will enable real-time processing of periscope imagery with minimal latency and provides significant performance advantages in low signal to noise ratio scenarios.
Benefits: The successful completion of this research will result in the development of optimized software capable of detecting small targets along the horizon in periscope imagery. This will significantly improve 360-degree situational awareness for submarine fleets. Additionally the novel active learning approach to small contact detection can be applied to a number of challenging problems including aircraft detection and port security.

Return