Remote Sensing of the Environment : Rio Grande Delta Marshes

Classification of Tiocano HYMAP data

Having analyzed the CASI data, the same algorithms were applied to the HYMAP data from which we expect better results. As mentioned before, HYMAP data carries more information than CASI. Figures 1 and 2 show the classified images obtained from the Best Basis and Hierarchical algorithms respectively with the class colors given below.

Fig 1. Tio Cano (HYMAP) - Hierarchical Tree Classifier

Fig 2. Tio Cano (HYMAP) - Best Basis Classifier

Table 1: Confusion matrix (%), Best Basis Classification

Class	1	2	3	4	5	6
1	100	0	0	0	0	0
2	0	100	0	0	0	0
3	0	0	95.79	0	2.0305	0
4	0	0	0	98.864	1.0152	0
5	0	0	4.2105	1.1364	96.9543	0
6	0	0	0	0	0	100

Overall accuracy: 98.6%

Table 2: Confusion Matrix (%), Hierarchical Classification

Class	1	2	3	4	5	6
1	100	0	0	0	1.0152	0
2	0	100	0	0	0	0
3	0	0	100	0	0	0
4	0	0	0	99.773	1.0152	0
5	0	0	0	0.2273	97.9695	0.2688
6	0	0	0	0	0	99.7312

Overall accuracy: 99.58%

The accuracies of both methods are higher than the ones for CASI data with Hierarchical classifier performing slightly better. Comparing the two classified images, we observe that the water class is significantly reduced by the Hierarchical algorithm. Overall importance plot of the bands shows that the ones which contributed most are between 40 and 60, these are the infrared bands whose wavelengths are not covered by CASI data. We can conclude here that the extra spectral information provided by HYMAP helped us in obtaining a more accurate classification than the one for CASI. Classified images of the two data sets look very different, however this may also be due to the fact that they were acquired in different seasons ; HYMAP (Sept. 17 1999) CASI (April 22 1998).

Figure 3. Best basis weighted bands for Tio Cano HYMAP data.

Last Modified: Wed Apr 14, 1999
CSR/TSGC Team Web