Research For Single Face Image Super-resolution Reconstruction Based On Sparse Representation And Wavelet Transform

In the field of digital image, the image resolution is one of the key indexes to evaluate the image quality, and it is also a very important parameter in the practical application. Super resolution reconstruction by the software algorithm can effectively improve the image resolution, this technique does not change the original hardware equipment, and it costs low. At present, it is widely used in various fields, such as national security, health care and social life and so on. With the development and application of face recognition technology, the human face as the carrier of human computer interaction and information recognition has received extensive attention, and more and more people are studying the technology of super resolution reconstruction.Sparse representation, as a new representation of signal, has become a hot topic in the field of image processing in recent years, and the various image processing algorithms based on the theory are also emerging. This paper improved on the basis of super-resolution reconstruction method based on sparse representation, and designed single face super-resolution reconstruction method based on sparse representations and wavelet transform. Firstly, the combination of median filter and Wiener filter is used to reduce the noise in the image, which is used to reduce the image noise and reduce the quality of the reconstructed image; Second, in view of the sparse representation reconstruction process, the image artificial effect and the reconstruction time is long, the method proposed in this paper has made the important improvement. Low resolution face image is decomposed into low frequency and high frequency by wavelet transform. The sparse representation is used to reconstruct the high frequency part of the image. The algorithm reduces the amount of computation and improves the efficiency of super resolution reconstruction; Third, using iterative back projection method to reconstruct the high resolution face image error correction, maintain consistency of super resolution reconstruction of high resolution image and the input low resolution image, prevent pixel of error introduced in the reconstruction process is diffusion.The simulation results show that the proposed algorithm is superior to the traditional super-resolution reconstruction method in both visual effect and objective evaluation, and the reconstruction speed of the algorithm is greatly improved.