Laser & Optoelectronics Progress, Volume. 55, Issue 12, 122801(2018)
Hyperspectral Anomaly Detection Algorithm Based on Combination of Spectral and Spatial Information
Fig. 1. Illustration of calculating pixel's spectral anomaly index and spatial structure anomaly index
Fig. 2. Flow chart of SSAD algorithm
Fig. 3. Experimental data 1. (a) Grey-scale map of the 130th band image; (b) spatial distribution map of targets
Fig. 4. Experimental data 2. (a) Grey-scale map of the 30th band image; (b) spatial distribution map of targets
Fig. 5. Experimental data 3. (a) Grey-scale map of the 100th band image; (b) spatial distribution map of targets
Fig. 6. Comparison of detection results on data 1 with different anomaly detection algorithms. (a) RX algorithm; (b) CRD algorithm; (c) LSMAD algorithm; (d) SSAD algorithm
Fig. 7. Comparison of detection results on data 2 with different anomaly detection algorithms. (a) RX algorithm; (b) CRD algorithm; (c) LSMAD algorithm; (d) SSAD algorithm
Fig. 8. Comparison of detection results on data 3 with different anomaly detection algorithms. (a) RX algorithm; (b) CRD algorithm; (c) LSMAD algorithm; (d) SSAD algorithm
Fig. 9. Comparison of ROC curves for different anomaly detection algorithms. (a) Data 1; (b) data 2; (c) data 3
|
|
Get Citation
Copy Citation Text
Huihui Ju, Zhigang Liu, Yang Wang. Hyperspectral Anomaly Detection Algorithm Based on Combination of Spectral and Spatial Information[J]. Laser & Optoelectronics Progress, 2018, 55(12): 122801
Category: Remote Sensing and Sensors
Received: May. 18, 2018
Accepted: Jun. 15, 2018
Published Online: Aug. 1, 2019
The Author Email: Liu Zhigang (dennylzg@163.com)