【摘要】 针对微表情识别中深度子空间网络鲁棒性差和泛化能力弱的问题，提出了基于非线性深度子空间学习和光流计算的微表情识别方法。该方法通过引入核变换充分挖掘微表情中的情感信息，同时使用光流特征捕捉微表情的运动信息，提高识别的鲁棒性。在SMIC、SAMM、CASME和CASMEⅡ4个广泛使用的自发微表情数据集和3DB-combined复合数据集上的实验表明，所提方法识别性能优于MACNN、Micro-Attention等深度学习方法，在复合数据集上的准确率达到0.834 6。此外，在SMIC数据集上添加10%、20%、30%和40%的随机噪声块后，在不同噪声水平下的未加权F1分数均优于其他算法，表明该方法在微表情识别任务中的有效性和鲁棒性。更多还原

【Abstract】 To address the issues of poor robustness and weak generalisation in deep subspace network-based microexpression recognition, this paper proposes a novel method that integrates nonlinear deep subspace learning with optical flow computation. The method employs kernel transformation to comprehensively extract emotional features from micro-expressions while simultaneously utilizing optical flow characteristcs to capture subtle motion dynamics, thereby enhancing recognition robustness. Experimental validation is performed on 4 widely adopted spontaneous micro-expression datasets(SMIC, SAMM, CASME and CASME Ⅱ) as well as a composite dataset 3DB-combined samples. Results demonstrate that the proposed method outperforms existing deep learning algorithms, including MACNN and Micro-Attention, achieving a recognition accuracy of 0.834 6 on the composite dataset. Furthermore, after adding 10%, 20%, 30%, and 40% random noise blocks to the SMIC dataset, the method consistently maintains superior unweighted F1 scores compared to other algorithms. These findings substantiate its effectiveness and robustness in real-world micro-expression recognition scenarios.更多还原