Accession Number : ADA319296

Title :   Multi-Scale Discriminant Analysis and Recognition of Signals and Images.

Descriptive Note : Technical rept.,

Corporate Author : MARYLAND UNIV COLLEGE PARK CENTER FOR AUTOMATION RESEARCH

Personal Author(s) : Etemad, Kamran

PDF Url : ADA319296

Report Date : MAY 1996

Pagination or Media Count : 131

Abstract : This dissertation explores multiscale discriminant basis selection, as well as the improvement of classification reliability through context-dependent integration of soft decisions. These methods are applied to texture and radar signature classification, document image segmentation, and human face recognition.

Descriptors :   *SIGNAL PROCESSING, *IMAGE PROCESSING, *PATTERN RECOGNITION, DATA BASES, ALGORITHMS, NEURAL NETS, OPTIMIZATION, RESOLUTION, THESES, TARGET DISCRIMINATION, SYNTHETIC APERTURE RADAR, RADAR IMAGES, ADAPTIVE SYSTEMS, COMPUTER GRAPHICS, KNOWLEDGE BASED SYSTEMS, DATA COMPRESSION, TARGET CLASSIFICATION, RADAR SIGNATURES, DISCRIMINATE ANALYSIS.

Subject Categories : Cybernetics
      Active & Passive Radar Detection & Equipment

Distribution Statement : APPROVED FOR PUBLIC RELEASE