主持课题 [1] 国家资助博士后研究人员计划“面向数据驱动和自主感知的高光谱遥感图像分类方法研究”，24万，2023.12-2025.07，结项，主持 [2] 中国科学院特别研究助理资助项目“面向复杂场景的高光谱遥感图像高精度分类方法研究”，80万，2023.10-2025.07，结项，主持 [3] 福建省自然科学基金创青项目“基于混合驱动与多信息挖掘的高光谱遥感图像目标识别方法研究”，8万，2026.5-2029.4，在研，主持 代表性论文 [1] Liu Yang, Liu Dongxu（共同第一作者）, He Lin et al. Temperature-Adaptive Thermal-Photonic Metamaterials: Generalized Infrared Engineering via Anharmonic Phonon Self-Energy Datasets and Dimensionality-Augmented Neural Networks, Advanced Functional Materials, 2025. [2] Luo Binxin, Liu Dongxu et al. Progressive Transformer With Multimodality Adaptation for RGB-T Tracking, IEEE Transactions on Instrumentation and Measurement, 2025. [3] Luo Binxin, Liu Dongxu*（通讯作者） et al. HSD2Former: Hybrid-Scale Dual-Domain Transformer with Crisscrossed Interaction for Hyperspectral Image Classification, remote sensing, 2024. [4] Liu Dongxu et al. A Hybrid-Scale Feature Enhancement Network for Hyperspectral Image Classification, remote sensing, 2024. [5] MiaoZhengwei, Luo Hui, Liu Dongxu, Zhang Jianlin. Improving Visual Representations of Masked Autoencoders With Artifacts Suppression, IEEE Signal Processing Letters, 2024. [6] Liu Dongxu et al. A Decompressed Spectral-Spatial Multiscale Semantic Feature Network for Hyperspectral Image Classification, remote sensing, 2023. [7] Liu Dongxu et al. A Multiscale Cross Interaction Attention Network for Hyperspectral Image Classification, remote sensing, 2023. [8] Liu Dongxu et al. A Multibranch Crossover Feature Attention Network for Hyperspectral Image Classification, remote sensing, 2022. [9] Liu Dongxu et al. A Hybrid-order Spectral-Spatial Feature Network for Hyperspectral Image Classification, remote sensing, 2022. [10] Liu Dongxu et al. A Discriminative pectral-Spatial-Semantic Feature Network Based on Shuffle and Frequency Attention Mechanisms for Hyperspectral Image Classification, remote sensing, 2022. [11] Liu Dongxu et al. A Novel 2D-3D CNN with Spectral-Spatial Multi-Scale Feature Fusion for Hyperspectral Image Classification, remote sensing, 2021. 发明专利 [1] 基于双支路多视角空谱特征交互的多谱段目标识别方法弱键合强度脆性键合样品截面透射电子显微镜制备方法 [2] 基于多视角交互式神经网络的特征提取方法 [3] 基于运动特性补偿的拓展目标特征点跟踪方法 [4] 一种基于自适应通道交换的渐进式多模态目标跟踪方法 [5] 一种基于耦合-解耦特征增强的多模态目标跟踪方法 | 