Object Discrimination via Bulk Temperature/Velocity Filtering. Phase 1
Abstract
This report describes work on the velocity filter approach to target detection and tracking, including its application to kinetic kill debris rejection. We present results of theoretical studies deriving the optimum (maximum-likelihood) filter for detecting a target moving through a sequence of N passive sensor image frames with a known, assumed or estimated vector velocity in which the target and background observations are modeled as independent Poisson processes and where both uniform and nonhomogeneous backgrounds are considered. Also presented are the results of algorithm simulation experiments conducted with (a) digital imagery recorded by the 30-inch Quad-Camera optical telescope and (b) simulated target imagery. Finally, a study of alternative approaches to processor implementation is presented. The implementation study includes consideration of a fine-grained, massively-parallel associative processor. Keywords: Discrimination; Signal processing; Computer architecture; Bulk filtering; Image processing; Parallel processing; Associative processing; Algorithms.
Document Details
- Document Type
- Technical Report
- Publication Date
- Apr 12, 1988
- Accession Number
- ADA194533
Entities
People
- William B. Kendall