arXiv cs.AI· Meihua Zhou, Liping Yu, Xinyu Tong, Wai Kin Fung, Ruiguo Hu, Jiarui Zhao, Nan Wan·· 12 小时前AI 评分20
SpectralTrain：面向高光谱图像分类的通用训练框架
SpectralTrain: A Universal Framework for Hyperspectral Image Classification
AI 导读
SpectralTrain 是一个与架构无关的通用训练框架，通过将课程学习与基于 PCA 的光谱降采样结合，提升高光谱图像分类的训练效率。在 Indian Pines、Salinas-A 和新提出的 CloudPatch-7 三个基准数据集上，训练时间缩短 2-7 倍，精度仅有小幅至中等变化，具体取决于骨干网络。框架不依赖特定架构、优化器或损失函数，兼容经典与 SOTA 模型，代码已开源。
正文 · 原文
Abstract: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 this https URL.
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2511.16084 [cs.CV] |
| (or arXiv:2511.16084v4 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2511.16084 arXiv-issued DOI via DataCite |
Submission history
From: Meihua Zhou [view email]
[v1]
Thu, 20 Nov 2025 06:19:26 UTC (5,580 KB)
[v2]
Wed, 13 May 2026 07:10:51 UTC (1,593 KB)
[v3]
Fri, 29 May 2026 08:48:21 UTC (1,593 KB)
[v4]
Tue, 20 Oct 2026 11:27:41 UTC (5,483 KB)
来源：arXiv cs.AI · arxiv.org