A Unified, Multiresolution Framework for Automatic Target Detection and Recognition
Abstract
The basic objective of this research program is to pursue an integrated set of research problems associated with Automatic Target Detection and Recognition (ATD/R). Our research is unified in the sense that it involves the investigation of problems spanning the complete processing chain from sensor signal processing to image analysis to object recognition, thereby allowing us to understand how these individual processing stages interact and influence each other and to define new approaches that cut across established decompositions of the ATD/R problem. A second unifying theme of our work is the development of multiresolution methods. These range from the use of statistically optimal multiresolution algorithms that provide both extremely efficient procedures for image analysis and an explicit and rational basis for addressing noise/resolution tradeoffs to methods for multiresolution characterizations of object geometry and corresponding algorithms for extraction of multiresolution geometric features and for object recognition.
Document Details
- Document Type
- Technical Report
- Publication Date
- Aug 01, 1997
- Accession Number
- ADA332303
Entities
People
- A. S. Willsky
- J. H. Shapiro
- W. E. L. Grimson
Organizations
- Massachusetts Institute of Technology