摘要：以滨海县为研究对象，获取研究区域Landsat8OLI遥感影像，

摘要：以滨海县为研究对象，获取研究区域Landsat8OLI遥感影像，利用QUEST决策树的分类方法对影像进行分类。将植被覆盖指数（NDVI）、迭代自组织数据分析技术（ISODATA）非监督分类作为地学辅助数据因子添加到分类波段中，构建多源数据集进行不同特征的融合，处理目标类别之间的非线性关系。该方法灵活性大，总精度达91．045％，Kappa系数为0．851，取得了比普通方法更好的精度。精度的提高有助于解决滨海县复杂的规划、决策和管理等问题。<br><br>关键词：遥感；QUEST决策树；土地利用分类；滨海县

摘要：以滨海县为研究对象，获取研究区域Landsat8OLI遥感影像，利用QUEST决策树的分类方法对影像进行分类。 将植被覆盖指数（NDVI）、迭代自组织数据分析技术（ISODATA）非监督分类作为地学辅助数据因子添加到分类波段中，构建多源数据集进行不同特征的融合，处理目标类别之间的非线性关系。 该方法灵活性大，总精度达91． 045％，Kappa系数为0． 851，取得了比普通方法更好的精度。 精度的提高有助于解决滨海县复杂的规划、决策和管理等问题。<br><br>关键词：遥感；QUEST决策树；土地利用分类；滨海县

Abstract: Taking Binhai County as the research object, the Landsat 8oli remote sensing image of the research area is obtained, and the image is classified by using the classification method of demand decision tree. The unsupervised classification of vegetation coverage index (NDVI) and iterative self-organizing data analysis (ISODATA) is added to the classification band as the auxiliary data factors of Geosciences. The multi-source data sets are constructed to fuse different features and deal with the non-linear relationship between the target categories. The method is flexible, with a total accuracy of 91.045% and kappa coefficient of 0.851, which is better than the common method. The improvement of precision is helpful to solve the complicated planning, decision-making and management problems of Binhai County.<br>Key words: remote sensing; quest decision tree; land use classification; Binhai County<br>