基于遥感影像的城市森林分类提取及生态价值估算研究

以北京通州新城区城市森林为研究对象,提出一种基于信息论的决策树分类方法,以提取Aster遥感影像中城市森林平面量,改善传统分类方法仅利用影像光谱特征信息的不足,高分辨率影像和地面调查数据验证显示其分类精度达84.6%.结合分类影像图,引入城市森林生态价值评估模型(CITYgreen模型)估算出通州新城区森林的生态价值约为598.87万美元.研究表明,该方法适合大面积区域的城市森林分类提取及其生态价值估算.

Research on Extracting Urban Forest and Estimating Its Ecological Value Using Remote Sensing Image

Taking the urban forest in the new district of Tongzbou, Beijing as the study case, this paper presents a strategy for extracting of the urban forest from Aster images by the classification decision tree algorithm based on information theory. The classification decision tree algorithm takes advantage of not only the spectral features, but also the accessorial information such as texture features, DEM, etc Then, based on the high-resolution images and the ground survey data, the validation and accuracy evaluation of the algorithm is carried out, the result have suggested that the accuracy of classification based on decision tree algorithm is higher than others,which can obtain a lot of reasonable rules quickly and effectively. The last, in the light of the classification result and the CITYgreen model which parameters have been configured on the urban forest characters in the study area, the ecological and economic value of urban forests in the research region has been figured out that the whole value is about $ 5. 988 million. The research results show that the theory and practical techniques is feasible and the method adapts to classification of urban forest and the rapid estimation and analysis of its ecological value for large area coverage.