当前位置>主页 > 期刊在线 > 计算机技术 >

计算机技术2019年23期

数据驱动下人脸识别的技术预测研究
齐小英¹,陈欣²
(1. 河北大学 管理学院,河北 保定 071002;2. 广西民族大学 软件与信息安全学院,广西 南宁 530006)

摘  要:在数字化时代,图像已经成为社会信息来源的重要组成因素。人脸作为一种数字图像,人脸识别相比其他生物识别,具有便捷性、非接触性、应用广泛等优点。人脸识别已经从对“人脸”这个“面”的识别逐渐转变为对“人脸”这个“系统”的识别。本文在《2018 年人脸识别研究报告》的基础上,结合人脸识别新闻事件的语义分析结果预测其发展趋势。结果表明,人脸识别的未来发展趋势主要包括五个方面:(1)建立复杂性、多元化的人脸模型;(2)静默活体检测技术与3D 动态人脸识别相结合;(3)人脸识别应用趋于场景化;(4)人脸识别的市场规模逐渐扩大;(5)人脸数据治理以保护个人隐私。


关键词:数据驱动;人脸识别;人脸检测;技术预测



中图分类号:TP391.41         文献标识码:A         文章编号:2096-4706(2019)23-0076-03


Research on Data-driven Face Recognition Technology Prediction

QI Xiaoying1,CHEN Xin2

(1.School of Management,Hebei University,Baoding 071002,China;2.College of Software and Information Security,Guangxi University for Nationalities,Nanning 530006,China)

Abstract:In the digital age,image has become an important component of social information sources. Face recognition is a kind of digital image. Compared with other biometrics,face recognition has the advantages of convenience,non-contact,wide application and so on. Face recognition has gradually changed from face recognition to face system recognition. In this paper,based on the 2018 Face Recognition Research Report,combined with the semantic analysis results of face recognition news events,the development trend of face recognition news events is predicted. The results show that the future development trend of face recognition mainly includes five aspects:(1)Building complexity and diversification;(2)the combination of silent living detection technology and 3D dynamic face recognition;(3)the application of face recognition tends to be situational;(4)the market scale of face recognition is gradually expanding;(5)face data management to protect personal privacy.

Keywords:data driven;face recognition;face detection;technical prediction


参考文献:

[1] 左腾. 人脸识别技术综述 [J]. 软件导刊,2017,16(2):182-185.

[2] 卢文峰. 人脸识别研究技术发展综述 [J]. 电子世界,2017(17):97.

[3] 人工智能学家. 清华Aminer 发布最新2018 人脸识别研究报告 [EB/OL].(2018-10-12).https://blog.csdn.net/cf2SudS 8x8F0v/article/details/83034545.

[4] 苏光大. 人脸识别在社会公共安全领域的应用 [J]. 中国安防,2015(14):12-14.

[5] SIROVICH L,KIRBY M. Low-dimensional procedure for the characterization of human faces [J]. Journal of the Optical Society of America. A,Optics and image science,1987,4(3):519-24.

[6] OJALA T,PIETIKÄINEN M,HARWOOD D. A comparative study of texture measures with classification based on featured distributions [J]. Pattern recognition,1996,29(1):51-59.

[7] Fisher R A. The Use of Multiple Measurements in Taxonomic Problems [J]. Annals of Human Genetics,1936,7(7):179-188.


作者简介:

齐小英(1997.04-),女,汉族,陕西西安人,硕士研究生在读,研究方向:社会信息处理、情报分析;

陈欣(1995.03-),女,汉族,广西南宁人,本科,研究方向:信息系统、信息安全。