###
DOI:
本文二维码信息
基于OpenPose-GLCM-hash的配电安监数据清洗技术
(1.广东工业大学;2.广东电网有限责任公司;3.广东顺畅科技有限公司)
Distribution safety monitoring data cleaning technology based on OpenPose-GLCM-hash
CHENG Liang-lun1, PEI Qiu-gen2, YAN Yu-ping2, YU Zi-yong2,3,1,3,4,3, SHAO Yan-yu2, RUAN Wei-cong2, YAN Qiang4, LIN Jia-xin2, CEN Chao-yu4, LI Guo-guang4
(1.Guangdong University of Technology;2.Guangdong Power Grid Company Limited;3.China;4.Guangdong Smooth Technology Company Limited)
摘要
相似文献
本文已被:浏览 5次   下载 0
投稿时间:2024-09-23    修订日期:2024-11-19
中文摘要: 随着现代配电网络规模和复杂性的不断增长,安全监控面临着严峻的挑战,其中数据管理和冗余问题尤为突出。本文提出了一种OpenPose-GLCM-hash数据清洗技术,结合灰度共生矩阵(Grey Level Co-occurrence Matrix,GLCM)的全局纹理特征和改进的OpenPose算法的人体姿态特征,通过这两类特征的加权组合生成图像哈希码,实现图像去重。实验结果表明,OpenPose-GLCM-hash算法在不同环境下具有高去重率和低误判率,尤其在噪声环境中表现出色。该研究为配电安监作业的数据管理提供了高效解决方案。
Abstract:With the increasing scale and complexity of modern distribution networks, security monitoring is facing serious challenges, especially data management and redundancy. In this paper, an OpenPose -GLCM-hash data cleaning technology is proposed, which combines the global texture features of GLCM and the human pose features of improved OpenPose algorithm, and generates an image hash code through the weighted combination of these two features to achieve image de-weighting. The experimental results show that OpenPose -GLCM-hash algorithm has high weight removal rate and low error rate in different environments, especially in noisy environments. This research provides an efficient solution for data management of power distribution safety supervision.
文章编号:20240923001     中图分类号:    文献标志码:
基金项目:
引用文本: