摘要
随着工业物联网的快速发展,异常数据对工业物联网的正常运作会产生巨大的损害。尽管已有很多相关研究,但异常检测工作仍然存在许多难以解决的问题。本文提出了一种高效且鲁棒的半监督异常检测胶囊网络和一种新颖的异常得分机制来进行图像的异常检测工作。本文提出的方法可以精确识别并输出空间关系。在两个数据集上的实验结果表明,CN4AD有着出色的异常检测效果。
With the rapid development of the industrial internet of things,the abnormal data will cause great damage to the normal operation of the industrial internet of things.Although there existed many related researches,there are still some problems that are difficult to deal with anomaly detection work.This paper proposes an efficient and robust semi-supervised capsule network for anomaly detection(CN4AD)and a novel abnormal scoring mechanism to deal with images anomaly detection work.The proposed method can accurately identify and output the spatial relationships.Experiments on two datasets show that the CN4AD has outstanding performance of anomaly detection.
作者
蔡翔宇
CAI Xiangyu(College of Computer and Cyber Security,Fujian Normal University,Fuzhou,China,350117)
出处
《福建电脑》
2023年第1期1-5,共5页
Journal of Fujian Computer
关键词
异常检测
胶囊网络
异常得分机制
工业物联网
Anomaly Detection
Capsule Network
Abnormal Scoring
Industrial Internet of Things
