Scene Content Analysis Program - Phase II.
Abstract
This report summarizes the development of map matching techniques for precision guided reentry vehicles using measurements extracted from data taken by sensors with different operating characteristics and under different viewing geometries. The case of matching optical to side looking radar images is considered in detail. Map matching with hierarchical search was developed and practical algorithms implemented in matching scences with a wide variety of contents. Theoretical analyses, which were subsequently verified by experiment, indicated that applications of this method have made the difficult task of optical to radar map matching possible at a computational efficiency of 3 orders in magnitude higher than current map matching methods. Map matching with this method has also resulted in a high probability of a match at a relatively low probability of false fix. Also included are analyses to determine perpective, scaling and rotational parameters from data extracted from image set. Geometric transformation for the correction of viewing geometry can be made using these parameters. Intensity transformation using sensor-dependent parameters was also developed. Practical algorithms were implemented for the corrections of the intensity of several sets of radar images.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jul 15, 1977
- Accession Number
- ADA045624
Entities
People
- Ernest L. Hall
- Robert Y. Wong
- Werner Frei
Organizations
- University of Southern California