提出双选融合注意力机制，提升高光谱图像分类精度。
Dual Selective Fusion Transformer Network for Hyperspectral Image Classification
- 通过动态选择多尺度空间谱信息进行融合
- 在四个数据集上准确率最高达97.66%
- 适合需要精准识别地物的遥感应用
Transformer 在高光谱图像（HSI）分类中已取得良好效果。然而，现有 Transformer 模型在面对具有多样地物类型和丰富光谱信息的 HSI 场景时面临两大挑战：(1) 固定感受野无法适应不同 HSI 物体所需的上下文尺度；(2) 上下文融合中的无效自注意力特征影响模型性能。为此，本文提出一种新型双选融合 Transformer 网络（DSFormer），通过灵活选择并融合不同感受野的特征，实现空间与光谱上下文的联合建模，有效减少无关信息干扰，聚焦关键空间-光谱标记。具体地，设计了核选择融合变压器块（KSFTB），通过自适应融合多尺度空间与光谱特征，学习最优感受野，增强对多样化 HSI 物体的识别能力。同时引入标记选择融合变压器块（TSFTB），在空间-光谱自注意力融合过程中有策略地选择和组合关键标记，捕捉最核心的上下文信息。在四个基准 HSI 数据集上的大量实验表明，所提 DSFormer 显著提升了地物分类精度，在 Pavia University、Houston、Indian Pines 与 Whu-HongHu 数据集上分别达到 96.59%、97.66%、95.17% 和 94.59% 的总体准确率，相较之前最优模型分别提升 3.19%、1.14%、0.91% 和 2.80%。代码将公开于 https://github.com/YichuXu/DSFormer。
原文摘要 · Abstract (English)
Transformer has achieved satisfactory results in the field of hyperspectral image (HSI) classification. However, existing Transformer models face two key challenges when dealing with HSI scenes characterized by diverse land cover types and rich spectral information: (1) A fixed receptive field overlooks the effective contextual scales required by various HSI objects; (2) invalid self-attention features in context fusion affect model performance. To address these limitations, we propose a novel Dual Selective Fusion Transformer Network (DSFormer) for HSI classification. DSFormer achieves joint spatial and spectral contextual modeling by flexibly selecting and fusing features across different receptive fields, effectively reducing unnecessary information interference by focusing on the most relevant spatial-spectral tokens. Specifically, we design a Kernel Selective Fusion Transformer Block (KSFTB) to learn an optimal receptive field by adaptively fusing spatial and spectral features across different scales, enhancing the model's ability to accurately identify diverse HSI objects. Additionally, we introduce a Token Selective Fusion Transformer Block (TSFTB), which strategically selects and combines essential tokens during the spatial-spectral self-attention fusion process to capture the most crucial contexts. Extensive experiments conducted on four benchmark HSI datasets demonstrate that the proposed DSFormer significantly improves land cover classification accuracy, outperforming existing state-of-the-art methods. Specifically, DSFormer achieves overall accuracies of 96.59%, 97.66%, 95.17%, and 94.59% in the Pavia University, Houston, Indian Pines, and Whu-HongHu datasets, respectively, reflecting improvements of 3.19%, 1.14%, 0.91%, and 2.80% over the previous model. The code will be available online at https://github.com/YichuXu/DSFormer.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。