Tree-species-classification-based-on-the-combination-of-hyperspectral-and-airborne-LiDAR:Hyperspectral remote sensing images have high spectral resolution, but they can only provide two-dimensional spatial information, and some materials may have similar spectrum. In addition, hyperspectral data has high redundancy, and the classification accuracy is reduced due to the Hughes phenomenon. LiDAR can provide reliable three-dimensional data and forest canopy characteristics.The code mainly includes single tree segmentation, feature extraction, feature importance analysis, KNN and SVM classification. The method proposed in the article was verified, and satisfactory experimental results were obtained.The overall classification accuracy is over 85$\%$, which is about 10$\%$ higher than the classification accuracy of single hyperspectral data.
// repository documentation
Was this content helpful?
โ 0(0 ratings)
Recent Feedback
Download README
Do you want to download the README.md file for essfm-14?
