Answers are grounded in this paper and cite their sources.
In allusion to hyperspectral remote sensing image denoising problem, the article proposes an image denoising algorithm based on nonlocal low rand dictionary learning. The basic thought of the algorithm is to make use of the strong correlation among various wavebands of the hyperspectral remote sensing image and meanwhile combine the nonlocal self-similarity and the local sparseness of an image to improve denoising performance. Firstly, combine the strong correlation of waveband images, the nonlocal self-similarity and the local sparseness to establish nonlocal low rank dictionary learning model. Then, adopt iterative method to solve the model to obtain redundant dictionary and sparse representation coefficient. Finally, adopt redundant dictionary and sparse representation coefficient to recover the image. Compared with existing advanced algorithms, due to the adoption of such strong correlation among various wavebands of the hyperspectral image, the algorithm mentioned in the article can well reserve the detailed information of the hyerspectral remote sensing image and improve visual effect. Meanwhile, the test result has verified the effectiveness of the algorithm mentioned in the article.
No takes yet. Share an insight, caveat, or question.
Zhihua et al. (2015) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: