23.5 Endmember Extraction
The reflectance Cuprite data scene in Figure 1.12 was used for experiments by PBS with BP/ID-BD to investigate an application in endmember extraction where a threshold value ε for ID-BD was empirically set to 0.01 to remove correlated bands. After BP/ID-BD, only 95 spectral bands were retained. Table 23.1 tabulates the first 50 spectral bands after BP/ID-BD using 7 BP criteria: variance, SNR, skewness, kurtosis, entropy, ID, and negentropy. Interestingly, the interband correlation for this scene is clearly characterized by second-order statistics rather than by high-order statistics (HOS), where only a few bands are removed by ID-BD using three specific HOS criteria: kurtosis, ID, and negentropy.
Table 23.1 Seven BP criteria to produce the first 50 bands after BP/ID-BD with bands removed by ID-BD for Cuprite data by setting ε = 0.01.
| 50 BP/BD bands | Bands removed by ID-BD | |
| Variance | 87/78/98/189/75/94/96/95/67/123/124/122/120/125/117/114/129/112/54/131/46/40/187/108/133/102/106/25/20/188/18/141/143/142/145/140/146/144/139/147/16/138/149/137/150/151/152/15/153/154 | 85/88/86/89/84/91/80/90/92/83/82/79/93/81/99/97/77/76/100/74/73/72/101/71/70/69/68/121/66/65/119/64/118/63/127/126/128/62/116/115/61/113/59/60/58/57/55/56/130/111/53/52/51/50/110/49/48/47/41/44/43/45/42/39/38/109/37/36/132/35/34/33/135/32/134/107/31/29/30/28/27/26/103/104/24/23/105/22/21/19/17 |
| SNR | 68/26/89/20/18/53/76/39/16/15/14/114/120/112/117/124/122/12/123/125/11/129/9/97/131/108/96/8/94/95/7/133/6/5/106/149/ ... |
Get Hyperspectral Data Processing: Algorithm Design and Analysis now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
