Title | ||
---|---|---|
Classification of Tiangong-1 hyperspectral remote sensing image via contextual sparse coding |
Abstract | ||
---|---|---|
The hyperspectral remote sensing is one of the frontier techniques in the remote sensing research fields. Applying the sparse coding model to the hyperspectral remote sensing image processing is a hot topic in hyperspectral information processing. To improve the accuracy of hyperspectral image classification, we propose a classification method based on the spatial-spectral join-t contextual sparse coding. Firstly, a dictionary is obtained by training using samples selected from the ground-truth reference data. Then, the sparse coefficients of each pixel are calculated based on the learned dictionary. Afterward, the sparse coefficients are input to the classifier and the final classification result is obtained. The visible and near-infrared hyperspectral remote sensing image collected by Tiangong-1 in Chaoyang District of Beijing is used to evaluate the performance of the proposed approach. Experimental results show that the proposed method yields the best classification performance with the overall accuracy of 95.74% and the Kappa coefficient of 0.9476 in comparison with other classification methods. |
Year | DOI | Venue |
---|---|---|
2015 | 10.1109/ICMLC.2015.7340910 | 2015 International Conference on Machine Learning and Cybernetics (ICMLC) |
Keywords | Field | DocType |
Hyperspectral image,Remote sensing,Sparse coding | Reference data (financial markets),Computer science,Remote sensing,Cohen's kappa,Artificial intelligence,Classifier (linguistics),Computer vision,Full spectral imaging,Information processing,Pattern recognition,Neural coding,Hyperspectral imaging,Pixel | Conference |
Volume | ISSN | Citations |
1 | 2160-133X | 0 |
PageRank | References | Authors |
0.34 | 9 | 5 |