基于人脸识别技术的电视节目特定人物识别系统设计与实现

Design and Realization of TV Program Specific Character Recognition System Based on Face Recognition Technology

为做好主流舆论引导和正能量传播,电视节目中应避免出现传播负面思想与价值观的政治人物、公众人物,给社会舆论带来负面影响。基于深度学习的人脸识别技术将大大提高电视节目特定人物审核效率。本文介绍了基于卷积神经网络的人脸识别技术,应用该技术设计并实现了电视节目特定人物识别系统,对全国近千个电视频道进行自动识别,根据实测数据分析了影响人脸识别性能的因素,并介绍了运行半年来的识别结果与发展前景。

In order to become a force to spread positive energy and play a leading role in guiding public opinion,the appearance of politicians and public figures who spread negative thoughts and values should be avoided in TV programs,which will have a negative impact on public opinion.Face recognition based on deep learning will greatly improve the efficiency of specific character recognition in TV program.This paper introduces face recognition technology based on convolutional neural network(CNN),and uses this technology to design and implement a TV program specific character recognition system,which automatically recognizes specific characters in nearly 1000 TV channels in China.According to measured data,this paper analyzes the factors that affect the performance of face recognition,and introduces recognition results in the past six months and development prospects.