面向对象的沿海地区土地利用/覆被信息提取研究  

Object-oriented land use/cover classification in coastal area

作　　者：杨小艳 刘文璐[1] 李龙 陈龙乾[2] 陈龙高 YANG Xiaoyan;LIU Wenlu;LI Long;CHEN Longqian;CHEN Longgao(School of Geography, Geomatics and Planning,Jiangsu Normal University,Xuzhou 221116,China;School of Environment and Spatial Informatics,China University of Mining and Technology,Xuzhou 221116,China;Department of Geography, Earth System Science,Vrije Universiteit Brussel,Brussels 1050,Belgium)

机构地区：[1]江苏师范大学地理测绘与城乡规划学院,江苏徐州221116 [2]中国矿业大学环境与测绘学院,江苏徐州221116 [3]Department of Geography,Earth System Science,Vrije Universiteit Brussel,Bnissels 1050

出　　处：《测绘通报》2019年第6期89-91,95,共4页Bulletin of Surveying and Mapping

基　　金：国家自然科学基金(41601087; 41271121);江苏省大学生创新训练重点项目(201710320029);江苏高校优势学科建设工程项目

摘　　要：土地利用/覆被(LUC)可为土地资源领域相关研究提供基础数据.本文构建了面向对象的LUC分类方法,并以沿海特殊土地类型区连云港市为例,应用Landsat 8影像开展了实证研究。结果表明:①总体分类精度达到85.06%,总体Kappa系数为0.83,超过了0.7的最低允许判别精度;②该方法可以有效地减少研究区因南北部区域耕地植被覆盖度不同导致的错分现象,并可以用于盐田与滩涂信息的提取工作;③该方法既可为研究区土地利用相关研究提供符合精度要求的数据.也可为其他沿海地区进行土地利用/覆被信息提取工作提供参考和借鉴。Land use/cover ( LUC) information is a very essential dataset in supporting the related research and application.To extract the LUC information in the coastal area,an object-oriented based classification method is proposed and applied to Lianyungang city using Landsat 8 imagery.The results indicate that the accuracy of the classification exceeded the minimum allowed discriminant accuracy 0.7 with the overall accuracy 85.06% and the overall Kappa coefficient is 0.83.The methodology can reduce the misclassification of arable land due to the vegetation coverage between the southern plains and northern hills,and can be used to extract the tideland and salt-field in the area.The methodology can provide the LUC data with agreed accuracy for the land related research for Lianyungang city,and support the extraction of LUC information in other similar coastal areas.

关 键 词：土地利用/覆被(LUC) 面向对象 分类规则 沿海地区 遥感 

分 类 号：P237[天文地球—摄影测量与遥感]