首页|基于社会监控视频的雨量反演

基于社会监控视频的雨量反演

扫码查看
原文链接
  • NETL
  • NSTL
  • 万方数据
  • 维普
针对利用机器视觉算法估算雨量低准确率的问题,提出基于社会监控视频的雨量反演算法.首先利用降雨分类网络剔除无雨视频;其次引入交替方向乘子法(alternating direction method of multipliers,ADMM)提取降雨视频的前景信息,并通过语义分割和背景差分方法选取感兴趣区域(region of interest,ROI);然后构建以灰度变化和饱和度为特征的高斯混合模型(Gaussian mix-ture model,GMM)筛选ROI区域内的雨滴;最后依据透视成像关系计算雨滴尺寸,使用气象学Gamma模型反演降雨量.实验结果表明,本文降雨分类方法的准确率在MWD(multi-class weatherdataset)到达91.3％,在真实的数据集到达77.0％,雨量估算结果相比于现有方法更为准确.
Rainfall inversion based on social surveillance video
Aiming at the issue of low accuracy of rainfall estimation by the machine vision algorithm,a rainfall inversion algorithm based on social surveillance video is proposed.Firstly,the rainfall classifica-tion network is adopted to remove the no-rain video.Secondly,the foreground information of rainfall vid-eos is extracted by using the alternating direction method of multipliers(ADMM),and the region of in-terest(ROI)is chosen by semantic segmentation and background subtraction methods.Thirdly,a Gauss-ian mixture model(GMM)characterized by gray-scale change and saturation features is constructed to choose the raindrops in ROI.Finally,the raindrop size is calculated according to the perspective imaging relations,and the rainfall is inverted through the meteorological Gamma model.The experimental results show that the rainfall classification accuracy of the method reaches 91.3％in the multi-class weather dataset(MWD)and 77.0％in the real dataset,and the rainfall estimation results are more accurate compared with the existing methods.

machine visionrainfall inversionrainfall classificationregion of interest(ROI)Gaussian mixture model(GMM)

陈赞、许建华、梁卓然、胡德云、张涵骁、杨焕强

展开 >

浙江工业大学信息工程学院,浙江杭州 310014

杭州市气象局,浙江杭州 310051

机器视觉 雨量反演 降雨分类 感兴趣区域(ROI) 高斯混合模型(GMM)

国家自然科学基金国家自然科学基金浙江省自然科学基金杭州市农业与社会发展科研项目政府间国际科技创新合作项目

6200232761976190LQ21F020017202004A072019YFE0124800

2023

10.16136/j.joel.2023.11.0476

光电子·激光
天津理工大学 中国光学学会

光电子·激光

CSCD北大核心
影响因子：1.437
ISSN：1005-0086
年,卷(期)：2023.34(11)
  • 12