摘要
随着社会经济快速发展,电子商务行业竞争愈发激烈,这对电子商务企业提出了新的要求,为满足不断增长的客户需求和服务,本文提出了一种电子商务线下实体店人脸识别算法,首先,把智能化技术引入到人脸识别中,借助云端技术实现数据存取,创建了硬件环境支持,研究了人脸识别技术的基本思想和方法,将样例图像扭转到同一角度下,经特征提取后与查询图像匹配比较,实现对扭转的鲁棒性和不变性。其次,提出一种线性分析算法,实现人脸识别数据和销售预测数据拟合,对电商企业实体店客流影响下的数据关系做出实证分析。
With the rapid development of social economy,the competition in e-commerce industry is becoming more and more fierce,which puts forward new requirements for e-commerce enterprises. In order to meet the growing needs and services of customers,this paper proposes a face recognition algorithm for offline physical stores in e-commerce. Firstly,this paper introduces the introduction of intelligent technology into face recognition,realizes data access by means of cloud technology,and establishes hardware environment support. In addition,the basic idea and method of face recognition technology are studied in detail. The sample image is scaled to the same angle to extract features and matched with the query image,so as to achieve the invariability and robustness of torsion. Secondly,this paper makes an empirical analysis of passenger flow behavior through face recognition data by means of linear analysis and calculation method.
作者
孙林
单双
耿莹莹
杨锦
李保军
SUN Lin;SHAN Shuang;GENG Ying-ying;YANG Jin;LI Bao-jun
出处
《北京财贸职业学院学报》
2019年第2期37-41,50,共6页
Journal of Beijing College of Finance and Commerce
基金
"互联网+社区服务"平台的设计与建设研究课题(课题编号:SM201851638003)
北京市教育委员会社科计划一般项目的成果
关键词
智能化
人脸识别
客流分析
算法
Intelligence
Face recognition
Passenger flow analysis
Algorithm analysis