a pa pr m s
 
Feature Based Pattern Recognition - Sonar

Typical sidescan sonar images of shallow water mines from a standard NSWC-PC SAS database

 

fbpr_figure1

 

fbpr_figure2

 

fbpr_figure3

 

Comparison of ROC curves

fbpr_figure4

With funding from both the U.S. Navy's Office of Naval Research and the Raytheon Company, in combination with substantial IR&D support, Prometheus Scientists have used noncommutative group harmonic analysis to automatically detect and identify underwater mines in side scan sonar data.

This technique has been validated on real data and is ready for implementation in a prototype system as well as application in other arenas such as passive sonar, radar, EO/IR object detection and classification, cryptography, land mine detection, and to automatically identify anomalies in CAT and MRI image data. The underlying mathematics is presented in the monograph Group Filters and Image Processing (Psypher Press, 2003) written by two Prometheus Principal Scientists, Myoung An and Richard Tolimieri.

Prometheus research has resulted in mission-ready radar, EO and IR algorithms for IED detection. Further information is available, upon request, with the permission of the appropriate Government authorities.

 

 

 

© Prometheus Inc. All rights reserved. Web Privacy Statement