USA flag logo/image

An Official Website of the United States Government

Infrared Background Clutter Metrics

Award Information

Agency:
Department of Defense
Branch:
Army
Award ID:
37090
Program Year/Program:
1997 / SBIR
Agency Tracking Number:
37090
Solicitation Year:
N/A
Solicitation Topic Code:
N/A
Solicitation Number:
N/A
Small Business Information
OPTIMETRICS, INC.
3115 Professional Drive Ann Arbor, MI 48104-5131
View profile »
Woman-Owned: No
Minority-Owned: No
HUBZone-Owned: No
 
Phase 1
Fiscal Year: 1997
Title: Infrared Background Clutter Metrics
Agency / Branch: DOD / ARMY
Contract: N/A
Award Amount: $98,939.00
 

Abstract:

The ability of either a human observer or an automated sensor to recognize a target in an infrared scene depends both on the target and clutter content of the infrared scene. Metrics have been developed to quantify the difference between the target and its background, to be used as independent variables in the modeling of both observer performance and of automated seeker performance. We believe that a metric designed particularly to measure the information content added by the presence of a target will be monotonically related with the ability of any observer to detect that target, and hence should form a robust and useful measure of target detectability. We propose to build a correlation-free description of the background portion of the image under investigation, measure the information content of the background region, in bits/pixel, and then code the target region in a coding scheme based on the entropy of the background region. The result will be an absolute measure of the number of bits (information) added to the target/background image by the presence of the target. This quantity will serve as a very appropriate independent variable against which observer and seeker performance can be modeled. We will have developed a technique whereby the information content introduced by the presence of a target can be computed. The ease with which a target can be detected in its background should be monotonically related to the information content added by the presence of the target. Thus, we will have developed a powerful technique for modeling detection performance of a variety of sensors.

Principal Investigator:

George H. Lindquist
3139731177

Business Contact:

Small Business Information at Submission:

Optimetrics, Inc.
3115 Professional Dr Ann Arbor, MI 48104

EIN/Tax ID:
DUNS: N/A
Number of Employees:
Woman-Owned: No
Minority-Owned: No
HUBZone-Owned: No