Application of dualband infrared imagery in automatic target detection

摘要

Target detection and reconiton are two important modules in a ptyical automatic target recognition (ATR) system. Usually, an automatic target detector produces many false alarms that could inclur very poor recognition performance in the subsequent target rcognizer. Therefore, we need a good clutter rejector to remove as many clutters as posisible from the ouputs of the detector, before, feeding the most l;ikely target detections to the recognizer. We investigate the benefits of using dualband forrward-looking infrared (FLIR) image to improve the performance of a eigne-neural based clutter rejector. with individual or combined bands as input, we use either principal component analysis (PCA) or the eigenspace separation transform (EST) to perform feature extraction and dimensionality reduction. The transformed data is then fed to an MLP that proedicts the identity of the input, which is either a target or clutter. We devise and MLP training algorithm that seeks to miaximize the class separation at a given false-alarm rate, which does not necessarily minimize the average deviation of the MLP ouputs from their target values. Experimental results are presented on a dataset of real dualband images.