基于深度学习的人脸识别技术在烟草商业内部监管的应用  

Application of Face Recognition Technology Based on Deep Learning in Tobacco Commercial Internal Supervision

作　　者：古振猛 孙士日 GU Zhenmeng;SUN Shiri(Information Center,Anshan Tobacco Company,Anshan Liaoning 114001,China)

机构地区：[1]辽宁省烟草公司鞍山市公司信息中心,辽宁鞍山114001

出　　处：《信息与电脑》2023年第6期181-183,共3页Information & Computer

摘　　要：烟草商业公司承担着向卷烟零售客户销售卷烟的业务。因为烟草行业对于零售客户可以购买的卷烟数量和销售情况有一定限制,所以零售客户在收货时需要进行拍照确认。这些收集的照片将会长期保留在公司的服务器中,等待人工审核和确认。当前内部监管面临两个难题,一是同一个接货人为多家店铺代订货,二是接货人与之前保存记录的信息不一致。这些行为都是潜在的违规行为。设计系统通过深度学习算法对比历史图片中的人脸数据信息,在发现违规问题时及时提醒相关工作人员,有效提高了监管的效率和质量。Tobacco commercial companies undertake the business of selling cigarettes to cigarette retail customers.Because the tobacco industry has high requirements on the quantity and sales of cigarettes that retail customers can buy,retail customers need to take photos to confirm when receiving goods.These collected photos will be kept in the company's server for a long time,and then manually reviewed and confirmed.There are two difficulties in internal supervision.One is that the same receiver may receive goods in different stores.The other is that the receiver is changed in the same store.There may be no timely cancellation and new handling of the license after the transfer of the store.These are potential violations.Through the comparison between the deep learning algorithm and the face data information in the historical pictures,it reminds relevant staff when violations are found,thus effectively improving the efficiency and quality of supervision.

关 键 词：机器学习 深度学习 卷积神经网络 人脸识别 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]