多个区域摄像头的人脸实时对比设计

Real-time Face Comparison Design of Multiple Area Cameras

提出了一种检测不同区域内的相似人员并能准确找出的一种人脸对比算法。该系统通过调用区域内的多个摄像头,采用VGG模型实时采集人脸图片,并通过改进人脸相似度算法对比相似人脸。通过深入分析当前基于深度学习的人脸特征进行对比的特点,开展了大规模数据的特征对比实验,实验结果证明了人脸特征对比方法的实际应用价值。

In the paper,a face matching algorithm is proposed,that can be accurately identified for similar people in different regions.The system collects face images in real time by calling multiple cameras in the area,and compares similar faces through an algorithm.Through in-depth analysis of the characteristics of the current face features based on deep learning,the feature comparison experiments of largescale data are carried out.The experiment results show that the face feature comparison objective method has application value.