MBTI：一种用于基于基础模型的高光谱图像分类的多分支高效微调框架
MBTI: A Multi-Branch Efficient Fine-Tuning Framework for Hyperspectral Image Classification with Foundation Models
浏览论文内容
中文总结 AI 辅助
针对高光谱基础模型因传感器光谱带配置差异难以直接转移用于HSI分类的问题，提出MBTI框架，通过多分支预处理、LoRA模块及注意力融合模块，在保留光谱信息同时适应下游任务，实验证明其性能优越且可训练参数少。
中文摘要 AI 辅助
高光谱基础模型从大规模未标记数据中学习可转移的光谱-空间表示，为适应有限标记样本的下游高光谱图像（HSI）分类任务提供了有效范式。但传感器间光谱带配置差异大，直接模型转移困难，现有策略会丢弃有用光谱信息并削弱局部光谱连续性。为此提出MBTI框架，它能在保留全波段光谱信息的同时使高光谱基础模型适应下游分类任务。首先引入保持光谱连续性的多分支预处理策略，将原始HSI分为多个连续光谱子集；其次在各分支插入独立的低秩适应（LoRA）模块；最后通过多分支通道注意力融合模块自适应地重新校准和整合所有光谱分支的特征。在三个公共高光谱数据集上的实验表明，MBTI与代表性分类方法相比具有竞争力和卓越性能，最终秩为8的配置下，仅约2.33% - 2.36%的参数可训练。
英文摘要
Hyperspectral foundation models learn transferable spectral-spatial representations from large-scale unlabeled data. They provide an effective paradigm for adapting to downstream hyperspectral image (HSI) classification tasks with limited labeled samples. However, spectral band configurations vary substantially across sensors, which makes direct model transfer difficult. Existing adaptation strategies often compress, select, or reshape the original spectra to match model-specific input requirements. These operations may discard useful spectral information and weaken local spectral continuity. To address this problem, we propose MBTI, a Multi-Branch efficient fine-tuning framework for Hyperspectral Image classification. MBTI adapts hyperspectral foundation models to downstream classification tasks while preserving full-band spectral information. First, we introduce a spectral-continuity-preserving multi-branch preprocessing strategy. The original HSI is divided into multiple continuous spectral subsets, and a band reuse mechanism is used when the remaining bands cannot form a complete branch. This avoids invalid padding and unnecessary spectral loss. Second, independent Low-Rank Adaptation (LoRA) modules are inserted into each branch. They enable different spectral intervals to learn task-specific discriminative features while keeping most pre-trained parameters frozen. Finally, a multi-branch channel attention fusion module adaptively recalibrates and integrates features from all spectral branches. Experiments on three public hyperspectral datasets show that MBTI achieves competitive and superior performance compared with representative classification methods. Under the final rank-8 configuration, only about 2.33\%--2.36\% of the parameters are trainable. The code will be available at https://github.com/Azhenmiddleblock/MBTI/tree/main.
发表机构
- School of Computer Science, Wuhan University（武汉大学计算机科学学院）
- State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University（武汉大学测绘遥感信息工程国家重点实验室）
- China Petroleum Pipeline Engineering Corporation（中国石油管道局工程有限公司）
- Hebei Key Laboratory of Underground Energy Storage Technology（河北省地下储能技术重点实验室）
- No. 7 Oil Production Plant, Changqing Oilfield Branch, PetroChina Company Limited（中国石油天然气股份有限公司长庆油田分公司第七采油厂）
- Faculty of Electrical Engineering and Computer Science, Ningbo University（宁波大学电气工程与计算机科学学院）
- College of Computer Science and Technology, National University of Defense Technology（国防科技大学计算机科学与技术学院）
- School of Aerospace Information, Space Engineering University（航天工程大学航天信息学院）
- Key Laboratory of Intelligent Processing and Applicati（智能处理与应用重点实验室）
机构由 AI 辅助整理，请以论文原文为准。
