About this document
SAM-GFNet for Hyperspectral Segmentation by liu66666692 is a document available to read on EtoBox.
The article presents SAM-GFNet, a hierarchical network designed for hyperspectral image semantic segmentation that integrates generalized visual features from the Segment Anything Model (SAM) with task-specific spatial-spectral characteristics. This approach enhances land cover classification accuracy and demonstrates superior performance on large-scale hyperspectral datasets. The proposed method achieves overall accuracies of 79.60% and 86.92% on benchmark datasets, surpassing existing baseline methods.
- Author
- liu66666692
- Language
- EN