Skip to Main Content
The availability of similar satellite data products from multiple sensors has focused much attention on the issue of continuity across satellite data products from past, current, and future sensors. Hyperspectral datasets acquired over a variety of land cover types are extremely useful in attempting to resolve spectral differences in the global datasets from different sensors. The datasets from the Earth Observing 1 (EO-1) Hyperion sensor are very suitable for this purpose, as is airborne hyperspectral data. In this paper, we examine the possibility of translating vegetation index (VI) data between two sensors by using imagery from the Hyperion sensor and utilizing the vegetation isoline concept. The objectives of this paper are to introduce and test a VI translation technique, focused on the spectral differences associated with sensor spectral bandpass filters. The translation of global VI datasets from one sensor to another requires a methodology applicable over various land cover types and throughout the wide ranges in VI values. To meet these requirements, a technique is proposed that utilizes adjustable translation coefficients, based on an estimation of the leaf area index value relative to a numerical canopy model. The theoretical basis of the proposed translation algorithm is explained in terms of the vegetation isoline concept. Its performance was tested through a numerical experiment with a Hyperion image, focusing on the normalized difference vegetation index (NDVI) as a representative vegetation index. The results indicate the potential of the isoline-based translation technique for stable translation throughout wide ranges of NDVI values.