Polygon-based classification applied on the unitemporal SPOT4 XS, SPOT5 XS, IKONOS XS, QuickBird XS and QuickBird Pansharpaned (PS) images is described. The study site is an agricultural area located near Karacabey, Turkey covering an area of about 95 km2. The objective was to assess the effect of the spatial resolution on the polygon-based classification of agricultural crops. Both the post-polygon and pre-polygon classifications were carried out. In the post-polygon classification, first, the images were classified on per-pixel basis through a Maximum Likelihood classifier. Then, for each field, the model class was computed and the field was assigned the label of the model class. In the pre-polygon classification, first, the mean values were calculated for each field. Then, the per-pixel Maximum Likelihood Classification was carried out using the mean bands. The post-polygon classification of the SPOT4 XS and SPOT5 XS images produced an overall accuracy of 76,1% and 81,4%, respectively. The IKONOS XS image provided the highest overall accuracy of 88,6%. On the other hand, the QuickBird XS and QuickBird PS images provided the overall accuracies of 83,7% and 85,8%, respectively. For the pre-polygon classification, the overall accuracies of the SPOT4 XS and SPOT5 XS images were computed to be 65,2% and 69,8%, respectively. Similar to the post-polygon classification, the IKONOS image provided the highest overall accuracy of 81,8% while the SPOT5 XS image revealed slightly better results than the SPOT4 XS image. The overall accuracies of the QuickBird XS and PS images were found to be 78,6% and 82,1%, respectively.