The process of object selection by multispectral systems is analyzed. One example of such a system is an aircraft control device, created for the selection of concrete surfaces, such as runways, against natural backgrounds. The reliability of the object recognition process depends on the probability of object selection. The optimal choice of bands for a multispectral system, using pattern recognition theory approaches, can provide the maximum selection probability. A method for choosing and substantiating the number of multispectral system channels and their bands is proposed. It is shown that a two-spectral system is sufficient to solve the selection task. A catalogue of optimal channel bands is composed. The solidity of the results and recommendations is corroborated by a Monte-Carlo simulation of the object selection process. The results of the investigation can be used during the creation of aircraft control devices.
Concrete object recognition by multispectral systems
1998
9 Seiten, 2 Quellen
Aufsatz (Konferenz)
Englisch
Eigenviews for Object Recognition in Multispectral Imaging Systems
British Library Conference Proceedings | 2004
|Object Detection In Multispectral High Resolution Images
SPIE | 1988
|