FRMN - A Face Recognition Model Based on Convolutional Neural Network

5th Annual International Workshop on Materials Science and Engineering

Shang, Shifeng^1 ; Liu, Haiyan^1 ; Qu, Qiang^1 ; Li, Guannan^1 ; Cao, Jie^1

Department of Information and Communication, Armored Forces Academy of PLA Army, Beijing, China^1

关键词: Convolutional neural network;    Global feature;    Image features;    Local feature;    Recognition models;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/585/1/012101/pdf
DOI  :  10.1088/1757-899X/585/1/012101

来源: IOP

【 摘 要 】

A face recognition model based on Convolutional Neural Network was proposed in this paper. This model mixed local and global image features at the largest extent. By acquiring the good local feature from mouth, nose and mouth, it is great improvement to enhance the accuracy for face recognition. From the experiment results, this model can better abstract the global feature and show a great improvement in face recognition.