Target identification by means of adaptive neural networks in thermal infrared images

A generic method for target recognition is presented. The stress is put on the methods based on the neural networks and more specifically on the ART models (Adaptive Resonance Theory). This type of artificial neural networks (ANN) has the advantage to be unsupervised and adaptive: it is indeed able to acquire and to adapt its long term memory taking into account the context evolution. ART networks recognize very quickly classes which are already known, they further learn new images very fast. Two versions of ART are investigated: ART1 which only works with binary data and ART2 which is working with analog data. In practice, ART1 seems to need larger images than ART2 to achieve the same efficiency, but is obviously faster. A preprocessor has been developed whose output is invariant to translation, rotation and scale changes of the input. The most important feature of this preprocessor consists in its ability to preserve visual interpretation, which is not the case for the more classical methods using Fourier-like and log/polar transforms.