Anomaly Detection Algorithm Using Entropy-Based Band Selection and Morphological Operations for Hyperspectral Images

摘要

Anomaly detection is one of the challenging topics in hyperspectral image processing due to its high spectral resolution. In this paper, a novel hyperspectral anomaly detection method, called entropy and morphological algorithm, is proposed. This method consists of three steps. First, select a band containing rich information for anomaly detection using a novel band selection algorithm based on entropy. Second, extract the background pixels of the selected band by applying morphological operation. Then, detect the anomaly by removing the background pixels from the selected band. Experiments conducted on real hyperspectral data sets show that the performance of our proposed method is quite competitive in terms of detection accuracy and computation time.