摘要
针对传统的图像识别方法很难快速、准确地对考生进行识别从而验证其身份,文中详细地分析了卷积神经网络的原理及特性,提出一种基于多通道输入的稀疏卷积神经网络的考生识别算法,并与支持向量机及传统卷积神经网络进行比较,实验结果表明,该算法提高了考生识别的准确率,而且识别的速度大幅提高。
With the development of information technology, the face recognition technology is applied to various examinations,but the traditional image recognition method is difficult to identify the examinees quickly and accurately,and is uneasy to verify their identities. The principle and characteristics of convolutional neural network are analyzed in detail. An examinee recognition algorithm based on sparse convolutional neural network with multi - channel inputs is proposed,and compared with the algorithms based on support vector machine and traditional convolutional neural network. The experimental results show that the algorithm can improve the recognition accuracy and recognition speed of examinee significantly.
作者
赵树枫
周亮
罗双虎
柯立新
ZHAO Shufeng;ZHOU Liang;LUO Shuanghu;KE Lixin(University of Shanghai for Science and Technology,Shanghai 200433,China;Network and Information Center,Shanghai Municipal Educational Examinations Authority,Shanghai 200433,China;Information Center of the Shanghai Education Committee,Shanghai 200003,China)
出处
《现代电子技术》
北大核心
2019年第13期61-64,共4页
Modern Electronics Technique
基金
上海市教育委员会重点项目基金(Z2017364001)资助~~
关键词
考生识别
卷积神经网络
人脸识别
身份验证
多通道输入
方法比
examinee recognition
convolutional neural network
face recognition
identity authentication
multichannel input
method comparison
