| SINGLE IMAGE SUPER RESOLUTION VIA COUPLED SPARSE AND LOW RANK DICTIONARY LEARNING |
| Paper ID : 1129-GEOSPATIAL (R3) |
| Authors: |
|
Sanaz Sahebkheir *1, ali Esmaeily2, mohammad saba3 1Department of Geomatics Engineering, Faculty of Civil Engineering and Transportation, University of Isfahan, Isfahan, Iran 2Department of Surveying Engineering, Graduate University of Advanced Technology, Kerman, Iran 3Department of Radiology, Medical Science University, Kerman, Iran |
| Abstract: |
| Limitations in imaging systems and the effects of changes in sensing cause limitation in acquiring high resolution images such as satellite images and magnetic resonance imaging (MRI). Sparsity can reduce the noises and improve the resolution. Super resolution in medical and satellite imagery is essential because low resolution image analysis is very difficult. Sparsity techniques have significant influence on computer vision specially when the main objective is extracting the meaningful information. The success of sparsity is related to the nature of signals such as image and sound which are naturally sparse because they founded based on Wavelet and Fourier equations. In this research, we propose a method in order to recover high resolution image from the corresponding low-resolution counterparts of both MRI and satellite images. First, we propose a general framework for learning couple low rank and sparse principal feature representation. Joint optimization of the nuclear and L1 norms extracts the global low rank structure and the local patterns embedded in the image. In that case the reconstructed image will be more informative and matrix decomposition problem can recover a noisy observation matrix into an approximation of low rank matrix and a second matrix which contains some low dimensional structure. We assume that by removing the blur and noise from these images, they will be reconstructed in the highest quality. The proposed method was compared with a variety dictionary learning approaches which address super resolution problem, such as tensor sparsity, Generative Bayesian and TV based methods. We demonstrated the results of applied method on MRI and satellite images, showing both visual and psnr improvements. Dealing with complex data in best manner shows the robustness of the proposed method. |
| Keywords: |
| Image Super Resolution, Sparse Representation, Magnetic Resonance Imaging, Health Management, Remote Sensing, Denoising |
| Status : Paper Accepted (Oral Presentation) |
