发表机构
Qingdao University of Technology; Nanyang Technological University（青岛理工大学; 南洋理工大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究针对高光谱图像分类的局部性约束与高计算复杂度问题，提出结合多尺度CNN和Mamba的MSCM-net模型，经多组基准实验验证其在降复杂度的同时可实现先进分类性能。
AI 中文摘要
高光谱成像已广泛应用于遥感和工程领域，因此其分类方法的研究至关重要。尽管基于CNN和Transformer的方法已取得进展，但仍面临局部性约束和高计算复杂度的问题。为解决这些问题，我们提出了一种创新的高光谱图像分类模型MSCM-net。具体而言，首先，我们提出了结合多尺度CNN与Mamba的模型架构，该架构由多尺度特征提取模块（MCSE）和多个堆叠的Mamba块组成，融合了多尺度CNN的局部特征提取能力与Mamba的长序列建模优势。其次，所提出的MCSE模块由多尺度卷积和SENet构成，不同尺度的卷积核提取具有不同感受野的局部信息，增强了空间与光谱信息的融合；同时，SENet使模型能自动学习多尺度特征中各通道的重要性。此外，我们还提出了双分支特征聚合模块，可进一步有效提取并整合中心像素所含的光谱信息与周围像素的空间信息。我们的模型在三个广泛使用的基准数据集上进行了大量实验，实验结果表明，MSCM-net可在降低计算复杂度的同时达到先进的分类性能。
英文摘要
Hyperspectral imaging is widely used in remote sensing and engineering. Therefore, research on its classification methods is crucial. While CNN and Transformer-based methods have advanced, they still face locality constraints and high computational complexity. To address these issues, we propose an innovative hyperspectral image classification model, MSCM-net. Specifically, first of all, a model architecture combining multi-scale CNN and Mamba is proposed. It consists of a multi-scale feature extraction module (MCSE) and multiple stacked Mamba blocks, which integrates the local feature extraction capability of multi-scale CNN and the long sequence modeling advantage of Mamba. Secondly, the proposed MCSE module consists of multi-scale convolution and SENet. Convolution kernels of different scales extract local information with different receptive fields, enhancing the fusion of spatial and spectral information. Meanwhile, the SENet enables the model to automatically learn the importance of each channel in the multi-scale features. Furthermore, we also propose a dual-branch feature aggregation module, which further effectively extracts and integrates the spectral information contained in the central pixel and the spatial information in the surrounding pixels. Our model has undergone numerous experiments on three widely used benchmark datasets. The experimental results show that MSCM-net can achieve advanced classification performance while reducing computational complexity.
