This paper studies the urban environment using remote sensing data with high spatial resolution airphotos and hyperspectral data. By using the ample spatial structural information contained in the high spatial resolution data and the fine spectral information contained in the hyperspectral data

a classification method based on edge detection and spectral analysis of the image was explored to classify the urban cover types on the per block bases. The result indicates that this method is effective for the study of whole urban environment.