基于Census变换的智能型防雷接地设备缺陷可视化系统设计  

Design of defect visualization system for intelligent lightning protection and grounding equipment based on Census transform

作　　者：李瑾 LI Jin(Ezhou Electric Power Survey and Design Institute Co.,Ltd.,Ezhou 436000,China)

机构地区：[1]鄂州电力勘察设计院有限责任公司,湖北鄂州436000

出　　处：《电子设计工程》2022年第4期165-169,共5页Electronic Design Engineering

摘　　要：为提高防雷接地设备缺陷定位的精度,提出基于Census变换的智能型防雷接地设备缺陷可视化系统设计。采用视频跟随扫描与控制方法采集智能型防雷接地设备缺陷图像,结合CT和红外传感识别方法,分析设备缺陷的可视化特征;构建设备缺陷可视化图像的边缘纹理特征分析模型,利用图谱特征解析和差异性特征点标定方法,重构设备缺陷可视化图谱;在边缘纹理特征分析模型下,通过机器视觉识别技术检测图像的缺陷特征,进而采用Census变换进行智能型防雷接地设备缺陷可视化分析和视觉跟踪定位。仿真结果表明,采用该方法进行智能型防雷接地设备缺陷可视化识别的准确性较高,缺陷定位精度高达97.7%。In order to improve the positioning accuracy of lightning protection grounding equipment,the design of intelligent lightning protection grounding equipment defect visualization system based on Census transformation is proposed.Using video following scanning and control method to collect the defect image of intelligent lightning protection grounding equipment,combined with CT and infrared sensor recognition method,the visual characteristics of equipment defects are analyzed;the edge texture feature analysis model of equipment defect visualization image is constructed,and the equipment defect visualization atlas is reconstructed by using the spectrum feature analysis and difference feature point calibration method;the edge texture feature of the equipment defect visualization image is reconstructed.Under the eigen analysis model,the defect features of the image are detected by machine vision recognition technology,and then the defect visual analysis and visual tracking positioning of intelligent lightning protection grounding equipment are carried out by using Census transform.The simulation results show that the accuracy of visual identification of intelligent lightning protection grounding equipment defects is high,and the accuracy of defect location is as high as 97.7%.

关 键 词：Census变换 智能型防雷接地设备 缺陷可视化识别 红外传感技术 

分 类 号：TM726[电气工程—电力系统及自动化]