Estimativa da produtividade e determinação de zonas de manejo, em culturas de grãos, por meio de videografia aérea multiespectral

Abstract

Remote sensing has been investigated the generation of spatial variability maps in agriculture as a possible alternative. The objective of this work was to evaluate multispectral aerial digital videographic images, regarding their potential in the grain crops yield estimate and in the delineation of management zones. Yield maps were also used. They refer to the two study areas, cultivated with wheat, in the winter of 2001, and with soybean, in the summer of 2002. The images were geometrically rectified and radiometrically normalized. Simple linear regression was also performed between images and yield maps, pixel-to-pixel, with very different results for the two areas. The same happened when the degree of agreement between crops yield and image data was analyzed through the Kappa Index. Aerial multispectral videography has a good potential in the delineation of management zones in areas with large spatial variability.