基于WorldView-2影像数据对杭州西湖区绿地信息提取研究  

Extraction of the Urban Green Space Based on WorldView-2 Images in West Lake District of Hangzhou

作　　者：钱军朝 徐丽华[1,2] 邱布布 陆张维[1,2] 庞恩奇 郑建华[3] Qian Junchao Xu Lihua Qiu Bubu Lu Zhangwei Pang Enqi Zheng Jianhua(College of Environment and Resource, Zhejiang A ＆ F University, Lin＇an Zhejiang 311300, China Zhejiang Provincial Key Laboratory of Carbon Cycling in Forest Ecosystems and Carbon Sequestration, Zhejiang A ＆ F University, Lin＇an Zhejiang 311300, China College of Landscape and Architecture, Zhejiang A ＆ F University, Lin＇an Zhejiang 311300, China)

机构地区：[1]浙江农林大学环境与资源学院,浙江临安311300 [2]浙江农林大学浙江省森林生态系统碳循环与固碳减排重点实验室,浙江临安311300 [3]浙江农林大学风景园林与建筑学院,浙江临安311300

出　　处：《西南林业大学学报（自然科学）》2017年第4期156-166,共11页Journal of Southwest Forestry University：Natural Sciences

基　　金：浙江省自然科学基金项目(LY15D010006)资助;国家自然科学基金项目(E080201)资助;浙江省林学一级重中之重学科学生创新计划项目(201516)资助

摘　　要：以杭州市西湖区为例,根据研究区域地物在World View-2遥感影像的特征差异进行区域划分。在每个分区内采用不同的多尺度和方式进行分割,构建多层次结构,综合利用光谱、形状、纹理等特征变量;采用CART决策树分类算法,选择最优特征及节点阈值分区域对杭州市西湖区的植被绿地信息进行提取;采用Jeffries-Matusita(J-M)距离法,确定纹理窗口尺度并筛选纹理特征。结果表明:本研究利用可分离指数J-M距离法得到影像地物草地、农用地、灌木、乔木最佳纹理窗口尺寸分别为5×5、11×11、13×13、13×13,对纹理尺度的选择和纹理特征的降维极大地提高了信息提取的精度及效率;基于面向对象的CART决策树分类法的总体分类精度相比基于像元的最大似然法的精度从76.53%提高到88.56%,Kappa系数从0.711 7提高到0.862 3,绿地平均用户精度从72.73%提高到84.63%;同时比常规的面向对象的方法更快速灵活地确定分类特征及阈值,大幅度地提高了提取效率及精度。According to the difference of objects in the WorldView-2 imagery in West Lake District of Hangzhou, sub-regions were divided. Within each partition, different multi-scale segmentation was used and a hierarchical structure was built. To make a comprehensive utilization of spectrum, shape and texture features of variables, the CART （classification and regression trees） decision tree classification algorithm was constructed to select the optimal characteristics and thresholds for each sub-region to map the entire green space of West Lake District. To determine the texture window size and optimize the texture features, the method of J-M （Jeffries-Matusita） distance was used. The results showed that with the method of J-M distance, the texture window size of grassland, agricultural land, shrubs and trees was 5×5, 11×11, 13×13, 13×13, respectively. It greatly improved the precision and efficiency of information extraction for the selection of texture window size and dimension of texture features. Comparing with the maximum likelihood method classification based on pixel, the overall accuracy was increased from 76.53% to 88.56%, and the kappa coefficient was improved from 0.711 7 to 0.862 3, the average user accuracy of green space was also increased from 72.73% to 84.63%; Comparing with the conventional object-oriented method, the proposed method is more quickly flexible to determine features and thresholds, greatly improving the efficiency and accuracy of classification.

关 键 词：区域 城市绿地 信息 J—M距离 决策树 特征变量 

分 类 号：S757.4[农业科学—森林经理学]