SAR图像洪水淹没区语义分割方法研究
Research on Semantic Segmentation of Flood Inundation Areas in SAR Images
王杰 1黄本胜 2陈亮雄 2杨静学2
作者信息
- 1. 广东省水利水电科学研究院,广东,广州,510635;中山大学土木工程学院,广东,珠海,519082
- 2. 广东省水利水电科学研究院,广东,广州,510635
- 折叠
摘要
研究了合成孔径雷达(synthetic aperture radar,SAR)图像水体与非水体的后向散射特性,围绕样本自动标注与增强训练这两个关键问题,利用阈值分割、水文约束与马尔科夫随机场设计了自动标注算法,并将特征增强网络与嵌入式样本增强相结合,提出了一种在有限样本条件下的SAR图像水体语义分割方法.以广东省"22·6"北江特大洪水为实验案例,采用了潖江蓄滞洪区的GF-3影像为实验数据.通过实验证明,本研究提出的方法能够有效识别洪水淹没范围,总体分类准确率达到了92.6%左右.
Abstract
This paper studies the backscattering characteris-tics of SAR images of water and non-water,and focuses on the two key issues of automatic annotation and enhanced train-ing strategy.Threshold segmentation,hydrologic constraint and Markov random fields(MRF)are used in designing the automatic labeling algorithm using,with the integration of the feature enhancement network and embedded sample enhance-ment,resulting in a semantic segmentation method for SAR images with limited samples.In this study,the"22·6"Beiji-ang extreme flood is taken as an experimental case,and the GF-3 images of Pajiang River detention area are used as the experimental data.According to the experimental results,it is evident that the proposed method is capable of distinguishing water from non-water with 92.6%overall accuracy.
关键词
遥感/洪水淹没区/自动标注/增强训练/语义分割Key words
remote sensing/flood inundation area/automat-ic annotation/enhanced training/semantic segmentation引用本文复制引用
出版年
2024
NETL
NSTL
万方数据