Remote sensing images retrieval based on sparse representation

To retrieve remote sensing images quickly and accurately, we use the online dictionary learning algorithm to train an over-complete dictionary of query images and database images respectively. The trained dictionary is used as an image's feature description. Then we calculate the similarity between query image and database image with the image similarity evaluation algorithm which is based on the sparse representation of image features. Finally we retrieve the remote sensing images according to the descending order of similarity. The experimental results, given in Figs. 4, 5 and 6 and Table 1, and their comparison with the existing methods show preliminarily that our novel remote sensing image retrieval method based on sparse representation can accurately retrieve remote sensing images.