摘要
随着互联网的高速发展和人工智能时代的到来,越来越多从前必须由人脑完成的工作能够利用计算机技术来完成,而深度学习的出现更解决了传统机器学习算法在计算机视觉领域、自然语言处理领域表现不佳的问题,使机器也能够拥有准确感知图像和语音的能力。人脸识别是深度学习网络最常见的应用场景之一,具有自然、直接、方便的特点,且不需要检测对象配合,因此非常适合用于公共安全领域的风险检测。研究充分结合海关实际需求,搭建基于深度学习技术的人脸识别模型,提供对通关旅客进行实时风险甄别的解决方案,以及海关通关风险防控场景的理论参考,为后续深度学习技术在海关业务的研究提供支撑。
With the rapid development of the Internet and Artificial Intelligence, more and more tasks which must be completed by human brains can be completed using computer technology. The emergence of deep learning has improved the poor performance of traditional machine learning algorithms in the field of computer vision and natural language processing, which provides machines the ability to accurately perceive images and speech. Face recognition is one of the most common application scenarios for deep learning networks. It is natural, direct, convenient, and does not require the cooperation of detection objects, so it is very suitable for risk detection in the field of public safety. According to the needs of customs, we built a face recognition model based on deep learning technology, put forward a solution for real-time risk discrimination of customs passengers, and proposed a theoretical reference for the risk prevention and control of customs clearance. Also, our research provided supports for the follow-up study of deep learning in customs business.
作者
闫宇宁
苏晓伟
万振龙
YAN Yu-Ning;SU Xiao-Wei;WAN Zhen-Long(The National Information Center,General Administration of Customs,Beijing 100005)
出处
《中国口岸科学技术》
2020年第3期26-32,共7页
China Port Science and Technology
关键词
深度学习
人脸识别
神经网络
风险甄别
Deep learning
face recognition
neural network
risk discrimination
