Hyperspectral Anomaly Detection Using Deep Learning: A Review

Hyperspectral image-anomaly detection (HSI-AD) has become one of the research hotspots in field remote sensing. Because HSI’s features integrating image and spectrum provide a considerable data basis for abnormal object detection, HSI-AD huge application potential HSI analysis. It is difficult to effectively extract large number nonlinear contained using traditional machine learning methods, deep incomparable advantages extraction features. Therefore, been widely used shown excellent performance. This review systematically summarizes related reference based on classifies corresponding methods into performance comparisons. Specifically, we first introduce characteristics challenges faced by dealing with these problems. Then, classify HSI-AD. Finally, method compared several mainstream sets, existing are summarized. The main purpose this article give more comprehensive overview future work.