Hyperspectral imaging and target detection algorithms: a review.

Target detection is the field of hyperspectral imaging where the materials or objects of interest are detected from images captured by hyperspectral sensors. This methodology has gained much significance in the area of military and defense, exploration of minerals, monitoring the food quality, and medical science. Various challenges such as errors occurring in the data at the time of image acquisition, data redundancy, and separation of background from desired targets however still exist. So in this study, we have given a brief introduction to hyperspectral imaging and have reviewed various atmospheric corrections, dimensionality reduction, and target detection techniques that help in overcoming these challenges related to hyperspectral data and target detection. Many of the researchers have worked significantly in this area and have acquired desired results by overcoming these challenges. Our review analysis has been approached from the perspectives of the synthetic dataset, public repositories, and developments made in pre-defined algorithms. Further, to percept the scope of the designed algorithms, the achieved results are also presented and compared for further improvements. The review study is presented in Filtering based and Optimization-based detection algorithms. The comparative analysis of various papers states that focus on optimization-based detection algorithms needs to be explored as these have gained much-desired results. [ABSTRACT FROM AUTHOR]