摘要
针对传统油田巡检效果不好、效率很低的问题,提出利用多旋翼无人机进行管线巡检的建议,首先对管线所在区域进行拍摄,将获得的原始污油彩色图像分别转换到Lab颜色空间和YUV颜色空间,其次分别对Lab颜色空间进行OTSU阈值分割,对YUV颜色空间进行加权模糊熵分割,最后将两种分割算法得到的结果进行合并、滤波,得到最终的分割图像。仿真结果表明,该方法能够有效地进行管线巡检并发现该区域的污油,还可粗略计算污油面积。
In this paper, aiming at problem that traditional oilfield inspection is not effective and the efficiency is very low,the authors proposed conducting pipeline inspection by using multiple rotors UAV. First, the area where the pipeline is located is photographed, and the obtained original sump oil color image is converted to Lab color space and YUV color space respectively; secondly, Lab color space will be conducted OTSU threshold segmentation, and YUV color space will be conducted weighted fuzzy entropy segmentation; finally, the results obtained by the two segmentation algorithms are combined and filtered in order to obtain final segmentation image. The simulation results show that this method can effectively carry out pipeline inspection and find sump oil existing in this area, and also can roughly calculate area of sump oil.
作者
邢致恺
贾鹤鸣
邢国军
朱柏卓
张森
朱传旭
XING Zhi-kai;JIA He-ming;XlNG Guo-jun;ZHU Bai-zhuo;ZHANG Sen;ZHU Chuan-xu(School of Mechanical and Electrical Engineering,Northeast Forestry University,Heilongjiang Harbin 150040 China;The Second Oil Extraction Plant,Daqing Oilfield Co.,Ltd.,Heilongjiang Daqing 163414 China;The Eighth Oil Extraction Plant,Daqing Oilfield Co.,Ltd.,Heilongjiang Daqing 163414 China)
出处
《科技创新与生产力》
2018年第7期63-65,68,共4页
Sci-tech Innovation and Productivity
基金
黑龙江省研究生教育创新工程资助项目(JGXM_HLJ_2016014)
关键词
油田管线巡检
无人机
图像处理
分割算法
颜色空间
inspection of oilfield pipeline
unmanned aerial vehicle
image processing
partitioning algorithm
color space