摘要
基于黄河防凌远程视频监视系统现场采集到的视频影像,采用图像识别的方法提取黄河沿岸关键冰凌参数,包括密集度及流速。对于冰凌密集度的提取方法,是先将摄像设备取得的视频变为单帧图像,然后对单帧图像进行二值化处理,利用图像分割技术得到冰凌所在区域,最后统计得出冰凌的密集度;对于冰凌流速提取,是用摄像设备取得视频图像序列,利用稠密光流法标记各图像的特征点位置,追踪特征点,从而得到移动距离,最后根据时间间隔计算得出冰凌的流动速度。
The ice concentration and velocity are two key parameters for the Yellow River ice protection,which can be extracted by image rec⁃ognition method.The videos were field acquired by the remote video monitoring system of the Yellow River ice protection.The measurement of ice concentration had two steps.Firstly,the video was converted to single frame images.Then the single frame images were binarized.The ice regions were recognized by image segmentation and the ice concentration could be counted easily.For the measurement of ice velocity,the video image sequence was obtained by the camera equipment.Then the positions of each image’s feature points were marked by the dense op⁃tical flow method and the feature points were tracked to obtain the moving distance.Finally,the ice velocity was calculated according to the interval time.
作者
张宝森
李春江
孙凯
范钦涛
季顺迎
ZHANG Baosen;LI Chunjiang;SUN Kai;FAN Qintao;JI Shunying(Yellow River Institute of Hydraulic Research,Yellow River Conservancy Commission,Zhengzhou 450003,China;Department of Engineering Mechanics,Dalian University of Technology,Dalian 116023,China;Dizaosoft(Dalian)Limited Company,Dalian 116023,China)
出处
《人民黄河》
CAS
北大核心
2021年第2期41-44,48,共5页
Yellow River
基金
国家重点研发计划项目(2018YFC1508402-2)
国家自然科学基金资助项目(51679102,42041006)。
关键词
图像识别
冰凌密集度
冰凌速度
测量方法
现场采集
黄河
image recognition
ice concentration
ice velocity
measurement method
field acquisition
Yellow River
作者简介
张宝森(1965-),男,河南洛阳人,高级工程师,主要从事防汛减灾及冰凌监测研究工作E⁃mail:976129493@qq.com。
