Search Performance with PIN Successfully Predicts Search Performance with an Advanced Sensor
Abstract
In the experiments reported here, trained military observers first searched for tracked targets in approximately 100 synthetically generated Paint The Night (PTN) images. Months later, a different group of trained military observers searched for tracked targets using imagery obtained with a particular advanced sensor operating in the 8-12 mu band. In both experiments, the maximum amount of time t(sub L) observers were permitted to search an image was controlled and took on the values 3, 6, 9 and 12 seconds. The probability P(t) for the observers detecting the target as a function of time was measured for different values of t(sub L), for targets that varied continuously between easy and hard to find, and excellent agreement in search performance was observed when comparing PTN and imagery obtained in field trials utilizing an advanced thermal imager. The excellent agreement in search performance using PTN and field imagery, the ability to control target signature and backgrounds in PTN and the relatively low cost of search experiments done utilizing PTN imagery, make PTN an excellent tool for search model development.
Document Details
- Document Type
- Technical Report
- Publication Date
- Aug 01, 2002
- Accession Number
- ADP023105
Entities
People
- Max Lorenzo
- Melvin Friedman
- Richard Vollmerhausen
- Timothy C. Edwards