中圖分類號： TP391
文獻標識碼： A
DOI： 10.19358/j.issn.2096-5133.2022.01.002
引用格式： 毛伊林. 基于人臉識別的安全驗證技術研究[J].信息技術與網絡安全，2022，41(1)：10-17.

Research on security verification technology based on face recognition

Abstract： With the vigorous development of artificial intelligence and optical imaging devices, face recognition technology has been widely used based on advanced hardware foundation and deep learning algorithm. Especially in the financial field, it can efficiently complete face brushing login, payment, verification and other services, greatly saving human cost and user time cost, and stand out among many biometric technologies with the advantages of non contact，easy collection, convenience, stability and uniqueness. But its security has also become the biggest bottleneck in the application of face recognition, which may endanger biological information security and even national security. Therefore, it is urgent to improve the anti attack ability and security verification technology of face recognition technology. In vivo detection technology is widely used in the field of human-computer interaction and video surveillance, which can better solve this problem. This technology can effectively identify whether the face image is a real face, and combined with the face recognition system to improve the security and reliability. This paper summarizes and analyzes the existing in vivo detection technologies to verify the security of face recognition.

Key words : face recognition；in vivo detection technology；deep learning；image texture；multispectral