Data processing: measuring – calibrating – or testing – Measurement system in a specific environment – Electrical signal parameter measurement system
Reexamination Certificate
2001-06-07
2003-09-16
Barlow, John (Department: 2863)
Data processing: measuring, calibrating, or testing
Measurement system in a specific environment
Electrical signal parameter measurement system
C702S116000, C702S181000, C702S189000
Reexamination Certificate
active
06622100
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a hyperspectral analysis tool and more particularly to a method for assessing the performance (i.e. the probability of detecting a target) of a hyperspectral sensing system.
2. Description of the Prior Art
Various tracking systems are known in the art which detect and track objects or targets of interest. Examples of such tracking systems are disclosed in U.S. Pat. Nos. 4,465,940; 5,129,525; 5,323,987; 5,333,815; and 5,445,453 as well as “A System for the Processing and Analysis of Multi- and Hyperspectral Data” by Lurie et al,
SIG Technology Review
, Winter 1994, pages 43-58, hereby incorporated by reference. In particular, infrared, multispectral and hyperspectral systems are known. These sensors detect the spectral radiance of an object, and in particular, the reflected light intensity and wavelength reflected from an object of interest. Infrared systems are known to detect infrared radiation in a single band. However, there are problems with such infrared detection systems. For example, the infrared band may contain a relatively strong background signal which approximates or even exceeds the expected intensity level of the target of interest. Even though such systems utilize threshold detectors, the detection threshold must be set relatively close to the background clutter signal which can result in relatively low detection rates. Accordingly, multispectral and hyperspectral detection systems have been developed to overcome this problem. Such multispectral and hyperspectral detection systems operate on a plurality of spectral bands and are thus able to provide relatively higher detection rates. For example, a multispectral detector may operate at about 10 or more frequency bands while a hyperspectral sensor may operate at a 100 or more frequency bands. Examples of multispectral sensing systems are disclosed in commonly-owned U.S. Pat. Nos. 5,300,780; 5,479,255; and 5,528,037. Hyperspectral sensors are discussed in “A System for the Processing and Analysis of Multi- and Hyperspectral Data” supra.
Unfortunately, relatively simple countermeasures, such as camouflage, flares, or in-band sources, are known to reduce the effectiveness of multispectral sensors. As such, systems have been developed for optimizing the performance of multispectral sensors. For example, as set forth in commonly-owned U.S. Pat. No. 5,528,037, the integration time as well as the bands are selected to optimize the signal-to-noise ratio of a multispectral sensor. The system disclosed in the '037 patent is adapted to be utilized with a relatively small number of bands (i.e. 10 or less) each with large noise where the objective is detection of a single target against all other backgrounds. Even with such optimization, the performance of such multispectral systems still falls below the performance by hyperspectral sensing systems. This higher performance can be attributed to higher spectral resolution and ability to discriminate among subtle spectral differences. Thus, hyperspectral sensors are becoming in more demand. Unfortunately, there are no known systems suitable for assessing the performance of hyperspectral sensors. As such, hyperspectral sensing system performance has heretofore not been able to be assessed. Thus, there is need for providing a method for assessing the performance of a hyperspectral sensing system.
SUMMARY OF THE INVENTION
Briefly, the present invention relates to a system and method for assessing the probability of detection of a target of a hyperspectral sensing system. The system is adapted to calculate the probability of detection of targets based on various sensor parameters, atmospheric conditions, and a specified combination of targets and backgrounds for a given false alarm rate. The system may be executed, for example, on an IBM compatible PC to allow the user to optimize the hyperspectral sensor and subsequent signal processing to a particular set of backgrounds and targets. The sensor models, atmospheric models and target and background profiles are initially applied to the system in the form of databases. As such, the system enables the user to select among the various parameters to optimize a hyperspectral sensor and the subsequent signal processing for a particular set of parameters.
REFERENCES:
patent: 4465940 (1984-08-01), Graff et al.
patent: 5129529 (1992-07-01), Manyek
patent: 5300780 (1994-04-01), Denney et al.
patent: 5323987 (1994-06-01), Pinson
patent: 5333815 (1994-08-01), Sardanowsky
patent: 5445453 (1995-08-01), Prelat
patent: 5479255 (1995-12-01), Denny et al.
patent: 5528037 (1996-06-01), Whitsitt
patent: 5805742 (1998-09-01), Whitsitt
Lurie, et al. “A System for the Processing and Analysis of Multi- and Hyperspectral Data”SIG Technology Reivew, pp. 43-58, Winter 1994.
Bruno William M.
Slater Mark
Whitsitt Stephen J.
Barlow John
Cherry Stephen J.
Katten Muchin Zavis & Rosenman
Northrop Grumman Corporation
LandOfFree
Hyperspectral analysis tool does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Hyperspectral analysis tool, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Hyperspectral analysis tool will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3095834