A High Accuracy Land Use/cover Retrieval System

The effects of spatial resolution on the accuracy of mapping land use/cover types have received increasing attention as a large number of multi-scale earth observation data become available. Although many methods of semi automated image classification of remotely sensed data have been established for improving the accuracy of land use/cover classification during the past forty years, most of them were employed in single-resolution image classification. Due to the more heterogeneous spectral-radiometric characteristics within land use/cover units portrayed in high resolution images, many applications of traditional single resolution classification approaches have not led to satisfactory results. In this paper, we propose a fast adaptive content-based retrieval system of satellite images database using relevance feedback. Through our proposed system, we apply a super resolution technique for the Landsat-TM images to have a high resolution dataset. The human-computer interactive system is based on modified radial basis function for retrieval of satellite database images. To improve the accuracy of the system, we apply the backpropagation supervised artificial neural network classifier for both the low and high resolution datasets.