EVALUATION OF AIRBORNE VIDEO DATA FOR LAND-COVER CLASSIFICATION ACCURACY ASSESSMENT IN AN ISOLATED BRAZILIAN FOREST

This research was designed to evaluate the operational utility of airb orne bispectral video data for reconnaissance assessment of land-cover variability and to document the viability of the video data for class ification accuracy assessment. There are distinct logistical advantage s utilizing airborne video data in isolated and environmentally sensit ive regions where there is limited preexisting aerial photography and poor infrastructure and where ground accessibility is difficult and ex pensive. In a study of land-cover characteristic in Mato Grosso, Brazi l, video data provided important insights into the variability and tra nsitions in land-cover that could be used to identify targets for fiel d work and training areas for subsequent satellite image (Landsat TM) classifications. Comparisons of Landsat TM classification accuracy ass essments derived from the airborne video and standard color photograph y revealed that comparable results can be achieved at a high statistic al level of significance (0.01). These results demonstrated that the v ideo data can provide information for accuracy assessments equivalent to more standard photographic point-sample data acquisition missions w ith the added benefit of easily acquiring far more data. Results of th e supervised Landsat TM classification coupled with a topographic mode l produced an overall accuracy of 68% with a Kappa coefficient of 0.60 , based upon a priori classes identified through evaluation of the air borne data and a site visit.