PATTERN RECOGNITION AND DETECTION BY MACHINE
Abstract
Pattern recognition is considered generally, but emphasis is placed on the following points: optimum estimation of statistical parameters so as to minimize the probability of incorrect classification, non-Gaussian and non-stationary situations, pattern detection in a continuing time series and calculation of error probabilities. Some of the work is specifically directed toward the problem of radio station recognition. A design procedure for pattern recognizing machines is suggested which uses results from this report and other referenced sources.
Document Details
- Document Type
- Technical Report
- Publication Date
- Mar 08, 1963
- Accession Number
- AD0418387
Entities
People
- Arthur E. Laemmel
Organizations
- New York University Tandon School of Engineering