The land cover pattern and crops structure of oasis irrigation area of middle reaches in the Heihe River Basin were monitored using CASI aerial data with high spatial and spectral resolution as transects. We designed a hierarchical classification structure integrated by pixelbased classification and object-based classification to map land cover types and crop planting structure in this region. According to surveyed reference data about land cover and visual interpretation from high resolution imagery, the accuracy of the classification result about land cover and crops pattern from CASI transect data was evaluated, and the result showed that overall accuracy was 84.61%, Kappa coefficient was 0.8262. Compared with landsat TM/ETM+ land cover product in Zhangye in 2007 within flight transects, CASI aerial data with high resolution was able to effectively identify the trees, shrubs and various crops. Land cover pattern of oasis irrigation area and various crop proportions within cropland area were analyzed using CASI classification result. The result showed that almost 59.1% of oasis irrigation area of middle reaches in the Heihe River Basin was barren or builtup region, and vegetated region accounted for 39.8%. Cropland, trees and grassland accounted for 39.4%, 5.3% and 0.1% respectively, and corn was staple crop, accounting for 96.1% within cropland area. This study demonstrated that aerial remote sensing data with high quality and high spatial resolution was able to effectively monitor spatial heterogeneity of underlying surface in the basin, and offer high-resolution information about underlying surface types for study on eco-hydrological process.