Abstract
An automatic pattern recognition algorithm, consisting of event detection, feature extraction, and a decision process, was developed. The complete processing system was labeled SIGNET. The feature extraction aspect of SIGNET formulated the autoregressive (AR) coefficients from a hybrid adaptive AR algorithm in combination with a weighted linear threshold element into a linear prediction residual (LPR). The structure of the data sequence was then identified by extracting those LPR segments, which established event type boundary phenomena. SIGNET was evaluated on two sets of data. The first set was comprised of nine independent underwater transient sources. The percent of correct recognition of SIGNET in that evaluation ranged from 93 to 98. The second data set was from a strong impulsive seismic source. The percent of correct recognition was 97. The mathematical foundation of SIGNET was also used successfully as a basis for a multiple source identification technique. [Work supported by ONR/Environmental Sciences.]
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
