集美大学学报（自然科学版）2025，Vol.30Issue(2)：186-197,12.DOI:10.19715/j.jmuzr.2025.02.11
基于动态卷积的标签不确定性自学习预测分配算法的面部表情识别
Facial Expression Recognition Based on Self-Learning Label Prediction and Distribution Algorithm Based on Dynamic Convolutional for Label Uncertainty
摘要
Abstract
To solve the problem of uncertain factors such as noise,fuzzy labeling,and micro expressions affecting the dataset in the field of facial expression recognition,a label uncertainty self-learning prediction al-location algorithm is proposed.The algorithm consists of three core modules:1)A self-attention weighting module employing dynamic convolution to achieve fine-grained pixel-level attention mechanism,effectively re-ducing computational overhead;2)A regularized ranking module that optimizes the model's handling of un-certain samples through sample weight reranking and reallocation;3)A label reassignment module that cor-rects labels for low-weight samples,thereby improving overall prediction accuracy.Experimental validation demonstrates the algorithm's efficacy in mitigating the impact of label uncertainty,exhibiting outstanding per-formance on publicly available datasets such as RAF-DB and MMAFEDB.关键词
面部表情识别/标签不确定性自学习预测分配算法/动态卷积/抗噪神经网络Key words
facial expression recognition/label uncertainty self-learning prediction allocation algorithm/dynamic convolution/anti-noise neural network分类
信息技术与安全科学引用本文复制引用
杨远奇,蔡岱立,谢泽凌,江恩杰..基于动态卷积的标签不确定性自学习预测分配算法的面部表情识别[J].集美大学学报（自然科学版）,2025,30(2):186-197,12.基金项目
中国高校产学研创新基金——新一代信息技术创新项目"基于SDN的车联网流量预测算法研究"(2021ITA06004) （2021ITA06004）
福建省中青年教师教育科研项目"车联网IoV流量预测算法研究"(JAT210671) （JAT210671）
集美大学诚毅学院中青年项目"基于transformer的推荐算法研究"(c13019) （c13019）
