Comparative Assessment of Some Target Detection Algorithms for Hyperspectral Images

Target detection is of particular interest in hyperspectral image analysis as many unknown and subtle signals (spectral response) unresolved by multispectral sensors can be discovered in hyperspectral images. The detection of signals in the form of small objects and targets from hyperspectral sensors has a wide range of applications both civilian and military. It has been observed that a number of target detection algorithms are in vogue; each has its own advantages and disadvantages and assumptions. The selection of a particular algorithm may depend on the amount of information available as per the requirement of the algorithm, application area, the computational complexity etc. In the present study, three algorithms, namely, orthogonal subspace projection (OSP), constrained energy minimization (CEM) and a nonlinear version of OSP called kernel orthogonal subspace projection (KOSP), have been investigated for target detection from hyperspectral remote sensing data. The efficacy of algorithms has been examined over two different hyperspectral datasets which include a synthetic image and an AVIRIS image. The quality of target detection from these algorithms has been evaluated through visual interpretation as well as through receiver operating characteristic (ROC) curves. The performance of OSP algorithm has been found to be better than or comparable to CEM algorithm. However, KOSP out performs both the algorithms.Defence Science Journal, 2013, 63(1), pp.53-62, DOI:http://dx.doi.org/10.14429/dsj.63.376