RADIOMETRIC NORMALIZATION OF ASTER, LANDSAT TM AND ETM+ SENSORS FOR MULTI-TEMPORAL NORMALISED DIFFERENCE VEGETATION INDEX DATA QUALITY IN A DRYLAND WOODLAND RESERVE

Abstract

Images acquired at different times and from different sensors for multi-temporal assessment usually have different amounts of haze and dust in the atmosphere. These differences can mask real changes or make similar land cover appears to have changed. Thus, the use NDVI for multi-temporal assessment derived from multi-sensor satellite images require radiometric normalization. In this study, Landsat TM, ETM+ and ASTER products were used. Empirical scene normalization technique was used balancing the radiometric attributes of the products after geometric rectification and atmospheric correction. The results showed that empirical normalization via PIFs have significance positive influence on the quality of NDVI and it is capable further reducing noise in the image