Paper
27 June 1996 Automatic air-to-ground target recognition using LWIR FPAs
Jean-Louis Amadieu, Vincent Fraysse
Author Affiliations +
Abstract
The theoretical potential of optical sensors in terms of geometrical resolution makes them the ideal solution for achieving the terminal precision guidance of today's missiles. This paper describes such a sensor, working in the 8 to 12 micrometer spectral domain by using a 64 by 64 IRCCD focal plane array, and whose main mission is to recognize various types of armored vehicles within complex scenes that possibly include other vehicles of similar nature. The target recognition process is based upon a Bayesian approach and can be briefly described as follows: after a classical processing stage that performs the filtering and the multi- thresholding, the target recognition algorithm evaluates a similarity level between the objects, including the target, seen in the IR scene and the 'theoretical' target whose some mean, generic features have been implemented in a database. The surroundings of the target and its orientation in the IR scene are 'a priori' unknown. The similarity level is based on calculation of the Mahalanobis distance between the object features vector and the mean features vector of the model; this calculation involves a covariance matrix which is significant of the errors affecting the measured features and that in particular stem form the limited spatial resolution of the sensor, the detector noise and the sensor- to-target range estimation error. With respect to the sensor hardware, its main opto-mechanical characteristics as well as some electro-optics data are indicates; some examples of target acquisition in complex scenes involving different kinds of IR counter measures are also presented.
© (1996) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Jean-Louis Amadieu and Vincent Fraysse "Automatic air-to-ground target recognition using LWIR FPAs", Proc. SPIE 2744, Infrared Technology and Applications XXII, (27 June 1996); https://doi.org/10.1117/12.243482
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Sensors

Target recognition

Missiles

Target acquisition

Data modeling

Infrared sensors

Long wavelength infrared

RELATED CONTENT

Managing nuisance parameters
Proceedings of SPIE (August 24 2000)
Enhancing emotional-based target prediction
Proceedings of SPIE (April 11 2008)
Group assignment problem
Proceedings of SPIE (January 05 2004)

Back to Top