动态融合频域特征，提升多源遥感图像分类精度
Dynamic Frequency Feature Fusion Network for Multi-Source Remote Sensing Data Classification
- 设计动态滤波模块，在频域自适应学习特征过滤核
- 在两个基准数据集上准确率超越现有最优方法
- 适合需要融合高光谱、雷达与激光遥感数据的研究者
多源数据分类是遥感图像解译中关键但具挑战性的任务。现有方法在建模频域特征时对不同地物类型适应性不足。为此，我们提出动态频域特征融合网络（DFFNet），用于高光谱图像（HSI）与合成孔径雷达（SAR）/激光雷达（LiDAR）数据的联合分类。具体而言，设计动态滤波块，通过聚合输入特征在频域动态学习滤波核，并注入频域上下文知识。同时，提出谱-空间自适应融合块，通过通道混洗操作增强谱域与空间注意力权重的交互，实现全面的跨模态特征融合。在两个基准数据集上的实验表明，DFFNet在多源数据分类任务中优于现有先进方法。代码将公开于 https://github.com/oucailab/DFFNet。
原文摘要 · Abstract (English)
Multi-source data classification is a critical yet challenging task for remote sensing image interpretation. Existing methods lack adaptability to diverse land cover types when modeling frequency domain features. To this end, we propose a Dynamic Frequency Feature Fusion Network (DFFNet) for hyperspectral image (HSI) and Synthetic Aperture Radar (SAR) / Light Detection and Ranging (LiDAR) data joint classification. Specifically, we design a dynamic filter block to dynamically learn the filter kernels in the frequency domain by aggregating the input features. The frequency contextual knowledge is injected into frequency filter kernels. Additionally, we propose spectral-spatial adaptive fusion block for cross-modal feature fusion. It enhances the spectral and spatial attention weight interactions via channel shuffle operation, thereby providing comprehensive cross-modal feature fusion. Experiments on two benchmark datasets show that our DFFNet outperforms state-of-the-art methods in multi-source data classification. The codes will be made publicly available at https://github.com/oucailab/DFFNet.
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。