计算机工程与应用2019，Vol.55Issue(5)：105-111,7.DOI:10.3778/j.issn.1002-8331.1803-0419
新型LeNet-FC卷积神经网络模型算法的研究
Research on New LeNet-FC Convolutional Neural Network Model Algorithm
摘要
Abstract
Aiming at the problem of overfitting, slow convergence, local optimization and low recognition accuracy of existing Convolutional Neural Network(CNN)in face recognition training, a new LeNet-FC convolutional neural net-work model is proposed in this paper.By improving the network structure of the network layer, reducing the convolution kernel, and using the optimized Logarithmic Rectified Linear Uni(t L_ReLU)activation function, the recognition accuracy of the model in face recognition training reaches 99.85%.Simultaneously, a face recognition system is designed based on the LeNet-FC convolutional neural network model. The recognition accuracy of the system in the simulation test experiment with ORL face database reaches 96%.关键词
人工智能/人脸识别/卷积神经网络/结构改进/激活函数优化Key words
artificial intelligence/ face recognition/ convolutional neural network/ network structure improvement/ activa-tion function optimization分类
计算机与自动化引用本文复制引用
白创,陈翔..新型LeNet-FC卷积神经网络模型算法的研究[J].计算机工程与应用,2019,55(5):105-111,7.基金项目
黑龙江省自然科学基金(No.F2015022) （No.F2015022）
黑龙江省普通本科高等学校青年创新人才培养计划(No.UNPYSCT-2017149, No.UNPYSCT-2017175) （No.UNPYSCT-2017149, No.UNPYSCT-2017175）
佳木斯市2017年度重点科研课题(No.170032). （No.170032）
