无人驾驶中基于ResNet深度模型的人脸微表情识别算法  

Facial Micro-Expression Recognition Algorithm Based on ResNet Depth Model in Unmanned Driving

作　　者：蔡臻[1] 李锋[1] 魏楚强 CAI Zhen;LI Feng;WEI Chuqiang(School of Information,Guangdong Communication Polytechnic,Guangzhou 510650,China)

机构地区：[1]广东交通职业技术学院信息学院,广州510650

出　　处：《微处理机》2023年第1期36-39,共4页Microprocessors

基　　金：2022年广东省科技创新战略专项资金项目(攀登计划)(pdjh2022b0857);2022年广州市基础研究计划基础与应用基础研究项目(1715);2022年度广东省普通高校重点科研平台和项目(2022ZDZX1058)。

摘　　要：为解决由于人脸微表情存在局部特征少、不同情绪差异性小等特点而带来的特征提取难度大、表情识别率低等问题,以进一步提高人脸微表情识别精度,提出一种基于ResNet深度模型的人脸微表情识别算法。算法针对无人驾驶中的行人人脸表情识别场合,主要包括数据预处理及模型构建,在保证数据集统一性的同时,能够有效提高微表情的识别率。使用FER2013数据集对ResNet-50模型进行实际验证,并与ResNet-18的表现加以对比。本算法在实验中获得98.7%的准确率,优于ResNet-18,充分验证算法模型的有效性。In order to solve the problems of difficult feature extraction and low expression recognition rate caused by the lack of local features and small differences between different emotions in facial microexpressions, and to further improve the accuracy of facial micro-expressions, a facial micro-expression recognition algorithm based on ResNet depth model is proposed. The algorithm mainly includes data preprocessing and model construction, which can effectively improve the recognition rate of micro-expressions while ensuring the uniformity of data sets. The FER2013 data set is used to verify the ResNet-50 model,and the performance is compared with that of ResNet-18. The accuracy of the algorithm is 98.7% in the experiment, which is better than ResNet-18, fully verifying the effectiveness of the algorithm model.

关 键 词：ResNet模型 微表情 人脸识别 无人驾驶 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]