Classification of Hyperspectral Remote Sensing

Conclusion

Mapping of the land cover and its response to wetland management practices using remotely sensed data from the AVIRIS sensor is being investigated at the KSC complex. Relative to previous studies of the area based on Landsat TM, the increased spectral resolution of AVIRIS data is useful for the identification of wetland vegetation (all in the 90% range) as well as for discrimination between various upland vegetation types.

The work performed for this study has illustrated the need for feature extraction techniques which can lead to an accurate classification of hyperspectral imagery. While no classification technique is uniformly superior, many methods can yield excellent results (which will reduce the amount of post-processing) if meaningful inputs are used. For this particular test site, use of the MNF transform data in conjunction with the Markov Random Field classification algorithm yielded the best classification results.

The increased spectral resolution of the AVIRIS imagery substantially improves the discrimination and classification of both the wetland and upland vegetation compared to broad-band sensors such as Landsat TM.