基于改进Retinex-Net的电力系统网络入侵检测方法研究
Research on Power System Network Intrusion Detection Method Based on Improved Retinex-Net
崔庆雄1
作者信息
- 1. 云南农业职业技术学院,云南 昆明 655031
- 折叠
摘要
由分解网络、增强网络和图像重建构建而成的改进Retinex-Net神经网络模型,在电力系统网络入侵检测过程中,将图片从RGB空间域中转换至HSV空间域,利用V分量对其进行处理,之后在分解、强化、降噪的操作下,从RGB中输出增强后的图片,将其作为实现电力设备状态感知的重要条件,为后续的电力系统网络入侵检测奠定坚实的基础.以改进Retinex-Net为基础进行电力系统网络入侵检测方法研究,旨在通过利用Retinex-Net端到端的可训练低亮度图像,达到增强网络的目的,提高电力系统网络的故障检测能力,从而形成高效的电力系统网络入侵检测方法.
Abstract
In the process of power system network intrusion detection,the image is converted from the RGB space domain to the HSV space domain,and the V component is used to process it,and then the enhanced image is output from RGB under the operation of decomposition,enhancement and noise reduction,which is used as an important condition for the realization of power equipment state awareness,and lays a solid foundation for the subsequent power system network intrusion detection.Based on the improved Retinex-Net,the research on the power system network intrusion detection method aims to enhance the network and improve the fault detection ability of the power system network by using the end-to-end trainable low-brightness image of Retinex-Net,so as to form an efficient power system network intrusion detection method.
关键词
改进Retinex-Net/电力系统/网络入侵检测Key words
Improving Retinex-Net/Power Systems/Network Intrusion Detection引用本文复制引用
出版年
2024
NETL
NSTL
万方数据