摘要
为提升电气设备的检测效率和准确性,研究提出基于计算机视觉的电气自动化智能检测方法,通过图像预处理提高图像质量,利用计算机视觉算法对预处理后的图像进行特征识别。数据显示,自适应中值滤波后图像的信噪比为32.3 dB,图像结构相似性为0.91,均高于均值滤波和中值滤波。通过特征点数量和筛选匹配优化的加速稳健特征算法的识别准确率平均值为96.1%,匹配成功率平均值可达97.3%。基于计算机视觉的系统的平均准确率为94.5%,高于传统观察法和维修记录法。研究表明,基于计算机视觉的电气自动化智能检测方法在提高电气设备检测效率和准确性方面的潜力。该方法可以有效地应用于电气自动化领域,实现对电气设备状态的自动化监测和诊断。
In order to improve the detection efficiency and accuracy of electrical equipment,a computer vision based intelligent detection method for electrical automation is proposed.Image quality is improved through image preprocessing,and computer vision algorithms are used to recognize the features of the preprocessed images.Test data shows that the average signal-to-noise ratio of the image after adaptive median filtering is 32.3dB,and the average structural similarity of the image is 0.91,both higher than that of mean filtering and median filtering.The average recognition accuracy of the accelerated robust feature algorithm optimized by the number of feature points and screening matching is 96.1%,and the average matching success rate can reach 97.3%.The average accuracy of computer vision based systems is 94.5%,which is higher than traditional observation and maintenance recording methods.Research has shown that intelligent detection methods for electrical automation based on computer vision have the potential to improve the efficiency and accuracy of electrical equipment detection.This method can be effectively applied in the field of electrical automation,achieving automated monitoring and diagnosis of the status of electrical equipment.
作者
付锐
姚丹
FU Rui;YAO Dan(Xi'an Institute of Communications Engineering,Xi’an 710300,China)
出处
《自动化与仪器仪表》
2024年第11期227-231,共5页
Automation & Instrumentation
基金
西安交通工程学院中青年基金项目《基于计算机视觉的电气自动化智能检测方法研究》。
关键词
计算机视觉
图像处理
电气设备
自动化检测
特征匹配
computer vision
image processing
electrical equipment
automated testing
feature matching
