SpectralTrain: A Universal Framework for Hyperspectral Image Classification
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910271047467008 |
|---|---|
| author | Zhou, Meihua Yu, Liping Tong, Xinyu Fung, Wai Kin Hu, Ruiguo Zhao, Jiarui Wan, Nan |
| author_facet | Zhou, Meihua Yu, Liping Tong, Xinyu Fung, Wai Kin Hu, Ruiguo Zhao, Jiarui Wan, Nan |
| contents | Hyperspectral image (HSI) classification typically involves large-scale data and computationally intensive training, which limits the practical deployment of deep learning models in real-world remote sensing tasks. This study introduces SpectralTrain, a universal, architecture-agnostic training framework that enhances learning efficiency by integrating curriculum learning (CL) with principal component analysis (PCA)-based spectral downsampling. By gradually introducing spectral complexity while preserving essential information, SpectralTrain enables efficient learning of spectral -- spatial patterns at significantly reduced computational costs. The framework is independent of specific architectures, optimizers, or loss functions and is compatible with both classical and state-of-the-art (SOTA) models. Extensive experiments on three benchmark datasets -- Indian Pines, Salinas-A, and the newly introduced CloudPatch-7 -- demonstrate strong generalization across spatial scales, spectral characteristics, and application domains. The results indicate consistent reductions in training time by 2-7x speedups with small-to-moderate accuracy deltas depending on backbone. Its application to cloud classification further reveals potential in climate-related remote sensing, emphasizing training strategy optimization as an effective complement to architectural design in HSI models. Code is available at https://github.com/mh-zhou/SpectralTrain. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_16084 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | SpectralTrain: A Universal Framework for Hyperspectral Image Classification Zhou, Meihua Yu, Liping Tong, Xinyu Fung, Wai Kin Hu, Ruiguo Zhao, Jiarui Wan, Nan Computer Vision and Pattern Recognition Artificial Intelligence Hyperspectral image (HSI) classification typically involves large-scale data and computationally intensive training, which limits the practical deployment of deep learning models in real-world remote sensing tasks. This study introduces SpectralTrain, a universal, architecture-agnostic training framework that enhances learning efficiency by integrating curriculum learning (CL) with principal component analysis (PCA)-based spectral downsampling. By gradually introducing spectral complexity while preserving essential information, SpectralTrain enables efficient learning of spectral -- spatial patterns at significantly reduced computational costs. The framework is independent of specific architectures, optimizers, or loss functions and is compatible with both classical and state-of-the-art (SOTA) models. Extensive experiments on three benchmark datasets -- Indian Pines, Salinas-A, and the newly introduced CloudPatch-7 -- demonstrate strong generalization across spatial scales, spectral characteristics, and application domains. The results indicate consistent reductions in training time by 2-7x speedups with small-to-moderate accuracy deltas depending on backbone. Its application to cloud classification further reveals potential in climate-related remote sensing, emphasizing training strategy optimization as an effective complement to architectural design in HSI models. Code is available at https://github.com/mh-zhou/SpectralTrain. |
| title | SpectralTrain: A Universal Framework for Hyperspectral Image Classification |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2511.16084 |