Detecting Aircraft in Low-Resolution Multispectral Images: Specification of Relevant IR Wavelength Bands

Abstract

International audienceWe address the problem of detecting a stealth aircraft flying far away from an observer with limited visibility conditions using their multispectral signature. In such environment, the aircraft is a very low-contrast target, i.e. the target spectral signature may have a similar magnitude to the background clutter. Therefore, methods accounting only for the spectral features of the target, while leaving aside its spatial pattern, may either lead to poor detection statistics or high false alarm rate.We propose a new detection method which accounts for both spectral and spatial dispersions, by inferring level sets of the Mahalanobis transform of the multispectral image. This combines the approach of the well known RX detector with some elements of the level set methods for shapes analysis. This algorithm is in turn used to specify the wavelength bands which maximize an aircraft detection probability, for a given false alarm rate.This methodology is illustrated in a typical scenario, consisting of a daylight air-to-ground full-frontal attack by a generic combat aircraft flying at low altitude, over a database of 30000 simulated multispectral InfraRed Signature (IRS). The results emphasize that, in the context of aircraft detection, there is great interest in using multispectral IRS rather than integrated IRS, as long as the infrared bands are well chosen

A Compact Mid-Wave Infrared Imager System With Real-Time Target Detection and Tracking

“…It can recognize camouflage in the presence of strong electronic interference. The infrared imaging system has strong detection and recognition ability for stealth aircraft, drones, and other hidden target or target under strong light interference, which is more suitable for high-precision observation than radar [10]. Meanwhile, the infrared imaging system can collaborate with radar and radio to fill the blind spot of radar [11], [12].…”

Constrained-Target Band Selection With Subspace Partition for Hyperspectral Target Detection

“…However, the nanoscale spectral resolution of HSIs also brings some problems, such as high computational complexity and redundant information [20]. In fact, there are a large number of redundant bands or low-SNR (signal-to-noise ratio) bands in HSIs, and some previous studies [21][22][23][24] have confirmed that the detection of specific target can be achieved with a small number of representative bands. As a result, dimensionality reduction (DR) is considered to be a necessary preprocessing step to remove the data redundancy for target detection.…”