This paper assesses the effect of pansharpening process in classification of tropical crop and forest areas. Supervised classifications based on Support Vector Machine were adopted. Different pansharpening methods using bilinear interpolation technique have been used to merge very high spatial resolution Quickbird multispectral and panchromatic imagery. To develop this study, seven sub-areas were extracted and human segmentations data were created. The quantitative results based on the mean of Probabilistic Rand Index, Variation of Information and Global Consistency Error, computed for all sub-areas, showed similar results by using (0.92, 0.87, 0.87, 1.23, 0,2 respectively) and by not applying (0.93, 0.89, 0.86, 1.23, 0.21 respectively) pansharpening methods.