Super-resolution image reconstruction via patch haar wavelet feature extraction combined with sparse coding

This paper presents a new approach to single-image super-resolution reconstruction, based on patch haar wavelet feature extraction combined with sparse coding. The training sample set is constructed by image patches haar wavelet transform to extract the horizontal, vertical and diagonal high frequency component composition column feature vector. Then, we train a pair of learning dictionaries which have good adaptive ability by using joint training method. Learning dictionaries combined with sparse coding theory to realize the image super-resolution reconstruction. As the experiment results show, the new method has good performs for recovering the lost high frequency information, and has good robustness.