Abstract
In this paper, we propose the convolutional sparse coding based model in the wavelet domain for the task of single image super-resolution (SISR). The conventional sparse coding based approaches work on overlapping image patches and use the dictionary atoms to sparse code an image patch. Further, at the final stage, an overlap-add mechanism is used to get the final high-resolution image estimate. However, these algorithms fail to take into account the consistency present in the overlapping patches which limits their performance. We propose the use of wavelet integrated convolutional sparse coding approach where instead of dictionary atoms we utilize the convolution summations between the learned filters and their mappings for sparse representation based SISR. The use of wavelets is proposed owing to their unique directional and compact features. A pair of filters are learned along with a mapping function for each wavelet sub-band to exploit the consistency among patches. The proposed wavelet integrated convolutional sparse coding model helps capture useful contextual information. The proposed model is evaluated on publicly available datasets for different scale-up parameters. To show the efficacy of the proposed model we compare it with recent state-of-the-art algorithms. The visual results along with the quantitative ones indicate that the proposed model performs well for the tasks of super-resolution.
Highlights
The task in Single Image Super-Resolution (SISR) is to estimate the High Resolution (HR) image from a Low Resolution (LR) observation
For the task of SISR, we model the output of a degradation model as; l = (h d) ↓ s where h d is the convolution operation between the HR image h and the blur operator d, ↓ s is the down sampling with the scale s
As the wavelet sub-band images have a very useful property of redundancy, we can exploit this property by convolution-based sparse HR and LR filters and mapping learning
Summary
The task in Single Image Super-Resolution (SISR) is to estimate the High Resolution (HR) image from a Low Resolution (LR) observation. In [8] the authors proposed a unified spatial-temporal-spectral framework based on a deep convolutional neural network (STS-CNN). It is a learning based model for focusing on the rebuilding of remote sensing imagery contaminated through dead pixels and thick cloud. We propose the convolution sparse coding based model using wavelets for the task of SISR. We learn the HR filters to estimate the desired level-1 HR wavelet sub-band images using the level HR feature maps by simple convolutions. As the wavelet sub-band images have a very useful property of redundancy, we can exploit this property by convolution-based sparse HR and LR filters and mapping learning. The proposed model utilizes the convolutional sparse coding integrated with wavelet analysis for the task of SISR. The filter sizes are reduced which reduce the overall computation cost
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
