An efficient single image super resolution algorithm based on wavelet transforms
Sara Akbarzadeh, Hassan Ghassemian, Faeze Vaezi · 2015
In this paper we introduce a single image super resolution algorithm based on wavelet transforms. To compensate the information loss due to the down sampling in DWT, we used stationary wavelet transform (SWT). We used inverse stationary wavelet transform which does not increase the image size, but it can reconstruct the details better than the other conventional methods. We also add bicubic interpolation, which is not used in the other conventional methods. Finally, the estimated high resolution image has been rectified by the error back projection. Results show the proposed algorithm has a better performance comparing the state of the art algorithms in reconstruction. The speed of proposed algorithm is faster than the other methods.