摘要
水生植被分布情况、结构和演变趋势对湿地生态环境变具有重要的指示意义和科学研究价值。基于Sentinel-2遥感数据,综合应用光谱信息、水体植被指数、最佳指数法(Optimal Index Factory,OIF)计算的纹理特征,结合随机森林分类法,构建特征优化后的随机森林水生植被提取模型,对于桥水库进行水生植被提取。结果显示:该方法能有效的提取出水生植被,总体精度为93.22%,Kappa系数为0.91。进一步与最大似然和支持向量机(SVM)方法进行对比分析,结果表明本算法的总体精度分别提高了19.96%、8.53%,Kappa系数分别提高了0.25、0.11。基于水生植被全年提取结果,分析了于桥水库的水生植被年内变化,发现于桥水库水生植被在五月份最繁盛,随后逐渐消减,直至十月份基本消亡。实验表明:特征优化后的随机森林分类法在Sentinel-2影像水生植被提取中具有较好的适用性。
The distribution, structure and evolution trend of aquatic vegetation have important directive significance and with spectral information, water and vegetation index, and texture features calculated by the Optimal Index Factory(OIF)method, using random forest classification method, it constructs a feature-optimized random forest aquatic vegetation extraction model and extracts aquatic vegetation from the Yuqiao Reservoir. The results show that the method can effectively extract aquatic vegetation, the overall accuracy is 93.22%, and kappa coefficient is 0.91. Compared with the maximum likelihood and support vector machine(SVM) methods, the results show that the overall accuracy of the algorithm is improved by 19.96% and 8.53% respectively, and the kappa coefficient is improved by 0.25 and 0.11 respectively. Based on the annual extraction results of aquatic vegetation, the annual variation of aquatic vegetation in Yuqiao reservoir is analyzed.It is found that the aquatic vegetation in Yuqiao reservoir is the most prosperous in May, and then gradually decreases until October. The experimental results show that the random forest classification method after feature optimization has good applicability in the extraction of aquatic vegetation based on Sentinel-2 images.
作者
张佩莹
张方方
李俊生
谢娅
张兵
ZHANG Peiying;ZHANG Fangfang;LI Junsheng;XIE Ya;ZHANG Bing(College of Information Technology,Shanghai Ocean University,Shanghai 201306,China;Key Laboratory of Digital Earth Science,Aerospace Information Research Institute,Chinese Academy of Sciences,Beijing 100094,China;College of Resources and Environment,University of Chinese Academy of Sciences,Beijing 100049,China;China University of Geosciences,Beijing 100083,China)
出处
《生态科学》
CSCD
2023年第1期40-48,共9页
Ecological Science
基金
国家重点研发计划项目(2021YFB3901202)
河南省科学院重大科研聚焦项目(210101007)
国家自然科学基金(41701402)。
关键词
随机森林
特征优化
于桥水库
水生植被
变化趋势
random forest
feature optimization
Yuqiao Reservoir
aquatic vegetation
changing trend
