Automatic Target Recognition in Multispectral and Hyperspectral Imagery Via Joint Transform Correlation

In this chapter, we review the recent trends and advancement on automatic target recognition (ATR) in multispectral and hyperspectral imagery via joint transform correlation. In particular, we discuss the one-dimensional spectral fringe-adjusted joint transform (SFJTC) correlation based technique for detecting very small targets involving only a few pixels in multispectral and hyperspectral imagery (HSI). In this technique, spectral signatures from the unknown HSI are correlated with the reference signature using the SFJTC technique. This technique can detect both single and/or multiple desired targets in constant time while accommodating the in-plane and out-of-plane distortions. Furthermore, a new metric, called the peak-to-clutter mean (PCM), is introduced that provides sharp and high correlation peaks corresponding to targets and makes the proposed technique intensity invariant. This technique is also applied to the discrete wavelet transform (DWT) coefficients of the multispectral and HSI data in order to improve the detection performance, especially in the presence of noise or spectral variability. Detection results in the form of receiver-operating-characteristic (ROC) curves and the area under the ROC curves (AUROC) are used to show the performance of the proposed algorithms against other algorithms proposed in the literature. Test results using real life hyperspectral image data cubes are presented to verify the effectiveness of these proposed techniques.