APP下载

基于同态滤波和遗传阈值的憎水性自动检测方法研究

2019-08-13张广东张玉刚温定筠姚境王晓飞高立超郭陆

计算技术与自动化 2019年4期

张广东 张玉刚 温定筠 姚境 王晓飞 高立超 郭陆

摘   要:首先采用基于同态滤波技术的局部直方图均衡化算法和自适应中值滤波算法消除复合绝缘子憎水性图像的高频噪声。其次,鉴于憎水性图像中水珠引起的反光和透明等干扰,采用最大类间方差作为目标函数和遗传算法作为阈值的优化算法,获取了良好的分割效果。最后,将最大水珠区域的图像的面积比、形状因子、伸长度、7个不变矩共10个特征参数输入BP神经网络,对7个憎水性等级进行判定,结果表明训练准确率和测试准确率分别高达94%和90%。

关键词:憎水性自动检测;同态滤波;遗传阈值;BP神经网络

中图分类号:TP391.41                                         文献标识码:A

Investigation on Automatically Hydrophobic Detection Based

on Homomorphic Filtering and GeneticThreshold

ZHANG Guang-dong1 ZHANG Yu-gang2,WEN Ding-jun1,

YAO Jing3,WANG Xiao-fei1,GAO Li-chao1,GUO Lu1

(1. Cansu Electric Power Research Institute of State Grid,Lan Zhou,Gansu 730070,China;

2. Gan Su Electric Power Company of State Grid,Lan Zhou,Gansu 730010,China;

3. School of Electrical and Information Engineering,Hunan University,Changsha,Hunan 410082,China)

Abstract:The high-frequency noise of composite insulator hydrophobic image is firstly eliminated based on the local histogram equalization algorithm for homomorphic filtering and adaptive median filtering algorithm. Then,in view of the reflection and transparency caused by water droplets in the hydrophobic image,good segmentation effect is obtained through adopting maximum interclass variance as objective function and genetic algorithm as a threshold optimization algorithm. Finally,ten feature parameters,including area ratio,shape factor,extension degree,seven invariant moments,are placed into the BPNN,and the results indicate that the training accuracy and testing accuracyare as high as 94% and 90%,respectively.

Key words:automatically hydrophobic detection;homomorphic filtering;genetic threshold;BP neural networks

絕缘子憎水性检测方法主要有三种:喷水分级法、静态接触角法和憎水性指示函数法[1-5]。其中,第一种和第二种是传统的检测方法,仅适用于实验室内。憎水性指示函数法近年来发展迅速,其利用摄像技术和数字图像处理技术来判定绝缘子憎水性等级。它又可细分为多种方法,典型的有均熵法以及形状因子法[6-8]。由于其克服了传统检测方法的缺点,如测量过程复杂、实验环境受限、实验条件苛刻等,故引起了众多工程界和科学界的学者的关注。形状因子法在判据的形成过程中没有考虑拍摄角度和拍摄距离的影响,并且也没有考虑水迹和水珠形状等因素,因此准确率较低[9-11]。均熵法的判据形成过程不够明确,缺乏较强的理论依据,并且需要海量的经验数据[12-16]。……

登录APP查看全文