基于目标分割和景观格局特征的城市土地利用分类

The Application of Landscape Ecological Concepts and Object Segmentation to Land Use Classification

根据相同土地利用类型景观格局特征相似的原理,在传统遥感分类方法的基础上,结合景观生态学理论,建立了土地利用分类新方法;应用SPOT遥感图像提取了北京市五环内的居民用地和非居民用地类型,总分类精度达到了85.9%,Kappa系数为71.1%。本研究结合学科交叉的优势,为遥感技术应用和土地利用信息提取提供了新思路。

Land use in urban areas is crucial for urban land management decision -making, environment monitoring and urban planning. According to the landscape ecology concept that the landscape patterns within the same land use type are similar, this paper presents a new land use classification approach which integrates landscape characteristics and high- spatial resolution remote sensing data. Some key landscape metrics were selected to quantify the landscape patterns of different land uses. Then, the integration of SPOT image and landscape characteristics was applied to land use classification within the 5th Ring Road of Belling. The overall land use classification accuracy was 85.9% with Kappa parameter being 71.1%. The results show that the specific landscape patterns of different land use types would signifieantly contribute to improving land use classification, and could potentially be applied to other urban areas.