摘要
采用电阻率成像(ERT)进行污染场地检测是近年来的研究热点之一,对ERT图像的分析现阶段主要依靠人工处理,ERT探测图像中污染区域的自动识别是亟需解决的关键问题。针对此问题,该文提出了一种的ERT检测图像的污染区域自动识别方法,该方法主要包括4个步骤:(1)生成ERT灰度图;(2)采用OTSU算法计算分割阈值;(3)采用阈值进行图像分割;(4)判定污染区域。数值模拟实验研究结果表明,当污染区域的电阻率与背景电阻率差异较大时(即两者比值≤70%),该方法能够有效识别出污染区域,且污染区域的位置与范围与模型设定值相近;当两者差异较小时(两者比值为80%),直接采用OTSU算法进行图像分割,不能正确识别出污染区域,此时可采用均值滤波对ERT检测图像进行预处理再进行图像分割,能够有效识别出污染区域。
It is a hot topic to take contaminated sites survey using electrical resistivity tomography (ERT)in recent years. However,the pollution area analysis of ERT images still mainly depends on manual processing.The automatic identification of contaminated areas in ERT images is a key issue and should be solved urgently.Aiming at this problem,an identification method of soil pollution area is proposed.The main steps of the method include converting ERT data to gray scale images,calculating the threshold using OTSU algorithm,imaging segmentation using the calculated threshold,and determining the soil pollution area.A numerical simulation experiment for contaminated sites is conducted and the results show that the proposed method can effectively identify the soil pollution area when the difference between the resistivity of pollution soil and the background resistivity in the contaminated area is large,while the method fails when the difference between the two is small. For solving this problem,an averaging filter is used to preprocess the ERT detection image and then the contaminated area can be identified effectively.
作者
能昌信
王玉玲
张波
闫岩
王蒙
NAI Changxin;WANG Yuling;ZHANG Bo;YAN Yan;WANG Meng(School of Information and Electronic Engineering,Shandong Technology and Business University,Yantai 264005,China;School of Information and Electrical Engineering,Shandong Jianzhu University,Jinan 250101,China)
出处
《环境科学与技术》
CAS
CSCD
北大核心
2018年第12期138-143,共6页
Environmental Science & Technology
基金
国家自然基金(61503219
51708529)
山东省重点研发计划(2016GGX101005)
中央级院所基本科研业务专项重点项目(2016YSKY14)
关键词
电阻率法
污染场地
大律算法
污染区域识别
electrical resistivity tomography
contaminated sites
OTSU
soil pollution area identification
