基于融合CNN和LBPH的学生人脸识别方法

打开文本图片集

中图分类号：TP391.4 文献标识码：A 文章编号：2096-4706（2025）20-0051-04

Abstract：The student face recognitionmethod of Convolutional Neural Network （CNN）and Local Binary Pattern Histogram（LBPH）issudiedinispaper.FirstlyLBPHisusedtoextractthelocaltexturefeaturesofthefaces，andthnthe pre-trained CNN model isutilized toobtain DeepLearning feature vectors.Thecosine similarityalgorithmisusedtoquantify theface similarityafterfusing the twotypesoffeature vectors.Experimentsshow thatthe proposed methodhas highrecognition accuracyandeficiencyincomplex scenariosofcollges anduniversities，andissuitable forapplicationscenariossuchas monitoring system， academic affairs system and library management system.

Keywords： face recognition; CNN; LBPH; cosine similarity

0 引言

随着人工智能技术的快速发展，人脸识别技术在多个领域得到了广泛应用，尤其是在教育领域。（剩余4590字）