Constrained Energy Minimization Anomaly Detection for Hyperspectral Imagery via Dummy Variable Trick
| Autor: | Chein-I Chang |
|---|---|
| Rok vydání: | 2022 |
| Předmět: | |
| Zdroj: | IEEE Transactions on Geoscience and Remote Sensing. 60:1-19 |
| ISSN: | 1558-0644 0196-2892 |
| DOI: | 10.1109/tgrs.2021.3124412 |
| Popis: | Constrained energy minimization (CEM) has shown great success in subpixel target detection. This paper develops a dummy variable trick (DVT) to extend CEM to CEM anomaly detector (CEM-AD) and shows that CEM-AD also enjoys the same success in animation detection (AD). Its idea converts a known specific target signature d imposed on CEM to an unknown specific target signature to develop a signal-to-background ratio (SBR)-constrained energy minimization (SBR-CEM) as an unknown specified-target constrained energy minimization (UST-CEM) which serves as a liaison to derive the desired CEM-AD without prior knowledge of d. Surprisingly, the derived CEM-AD turns out to be a sample correlation matrix R-based anomaly detector, R-AD in correspondence to the well-known sample covariance matrix K-based anomaly detector developed by Reed-Xiaoli, RX-AD. To further improve CEM-AD, a low rank and sparse matrix decomposition (LRaSMD) model introduced by go decomposition (GoDec) and its sparsity cardinality (SC) are further incorporated into CEM-AD where two new versions of SC, fixed sparsity cardinality (FSC) and variable sparsity cardinality (VSC), are particularly designed to enhance AD. Finally, to effectively evaluate AD performance, recently developed 3D receiver operating characteristic (ROC) curve-derived detection measures are used for comparative studies and analyses. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|