Hyperspectral Image Super-Resolution via Recurrent Feedback Embedding and Spatial-Spectral Consistency Regularization
ABSTARCT :
Hyperspectral images with tens to hundreds of spectral bands usually suffer from low spatial resolution due to the limitation of the amount of incident energy. Without auxiliary images, the single hyperspectral image super-resolution (SR) method is still a challenging problem because of the high-dimensionality characteristic and special spectral patterns of hyperspectral images. Failing to thoroughly explore the coherence among hyperspectral bands and preserve the spatial-spectral structure of the scene, the performance of existing methods is still limited. In this article, we propose a novel single hyperspectral image SR method termed RFSR, which models the spectrum correlations from a sequence perspective. Specifically, we introduce a recurrent feedback network to fully exploit the complementary and consecutive information among the spectra of the hyperspectral data. With the group strategy, each grouping band is first super-resolved by exploring the consecutive information among groups via feedback embedding. For better preservation of the spatial-spectral structure among hyperspectral data, a regularization network is subsequently appended to enforce spatial-spectral correlations over the intermediate estimation. Experimental results on both natural and remote sensing hyperspectral images demonstrate the advantage of our approach over the state-of-the-art methods.
EXISTING SYSTEM :
? The spectral correlation, which is characterized by that there exists strong correlation among neighboring spectral bands of hyperspectral image, has been widely used for hyperspectral image reconstruction and analysis.
? When compared with perceptual and adversarial losses, which may restore details that do not exist in the original images and is undesirable in remote sensing field, l2 and l1 losses are more credible.
? Super-resolution reconstruction can infer a high-resolution image from one or sequential observed low-resolution images.
? It is a post-processing technique that does not require hardware modifications, and thus could break through the limitations of the imaging system.
DISADVANTAGE :
? In order to deal with the above problems, in this paper, we proposed a novel network for single hyperspectral super-resolution, namely Spatial-Spectral Feedback Network (SSFN).
? The differences in the number of spectral bands and imaging conditions will make it difficult to establish a unified deep network. Furthermore, it is easy to cause the over-fitting problem.
? . However, there are some drawbacks to these methods. Firstly, in the testing process, they typically need to solve complex and time-consuming optimization problems.
? Image super-resolution is a ill-posed problem, which calls for additional prior (regulation) to constrain the reconstruction process.
PROPOSED SYSTEM :
• To exploit the redundancy and correlation in spectral domain, some approaches have been proposed by exploiting the sparsity, non-local similarity, superpixelguided self-similarity, clustering manifold structure , tensor and low-rank constraints .
• Comprehensive ablation studies demonstrate the effectiveness of each component and the fusion strategy used in the proposed method.
• The proposed SSPSR method contains four main components including Grouping Strategy (GS), Progressive Upsampling (PU), Parameter Sharing (PS), and Spectral Attention (SA).
• In the proposed SSPSR method, in order to make the training process more efficient, we share the network parameters of each branch across all groups.
ADVANTAGE :
? We observe that with the help of local-spectral grouping strategy, the spectral reconstruction performance is greatly improved compared with the network without the grouping strategy (G=1).
? In order to obtain the best performance, we carefully adjust the hyperparameters of these comparison methods. The bicubic interpolation is used as our baseline.
? Through these approaches have achieved very good performance, the co-registered auxiliary image with high-resolution is very arduous, which limit the progress of practical application.
? Therefore, the performance of the super-resolution methods is heavily dependent on whether the designed prior can well characterize the observed data.
|