Study on super-resolution reconstruction algorithm based on sparse representation and dictionary learning for remote sensing image

Super-resolution image reconstruction plays a very important role in the interpretation of remote sensing images. Especially when the resolution of images is low, the size of the objects to be identified is close to the minimum resolution, and can be reconstructed by super-resolution better interpretation of the feature. In this paper, K-SVD algorithm is used to study the exampler of high resolution image library, and the dictionary of high resolution remote sensing image is obtained. The low resolution image is represented by high resolution dictionary, and the remote sensing reconstruction of remote sensing image is realized. Which improves the peak noise ratio and mean square error of the image, and has better performance than the interpolation algorithm. The method proposed in this paper has important significance and application prospect in remote sensing image application.