A Novel MRI image super-resolution reconstruction algorithm based on image representation and sparse dictionary learning

In this paper, we analyze the novel MRI image super-resolution reconstruction algorithm based on the image representation and sparse dictionary learning. At present, the high resolution image mainly by improving the precision of hardware devices such as optical devices and sensors to obtain while high precision hardware is expensive, however, people often want to in low economic cost under the premise of a higher resolution image. Multi-frame image super resolution reconstruction is based on pixel level and accurate motion estimation on the basis. Sparse coding is a kind of mammalian visual simulation system is the primary visual cortex simple cells receptive field method, it has been in the field of image processing and pattern recognition, etc. have made some progress. Our research combines the advances of the image representation and sparse dictionary learning to modify the traditional method with the theoretical optimization. The experimental result proves the effectiveness of the algorithm.