山东农业大学学报（自然科学版）2017，Vol.48Issue(1)：70-74,5.DOI:10.3969/j.issn.1000-2324.2017.01.014
密度分割和面向对象结合的东平湖湿地信息提取
Information Extraction from Dongping Lake Wetland with the Density Segmentation and Object-oriented Method
摘要
Abstract
Because wetland information has the characteristics of complexity and diversity in the lake region, it is difficult to acquire accurate information from remote sensing image. Selecting Dongping Lake wetland as the research object, using Landsat ETM+ image data, the paper extracted vegetation types accurately through density slicing and threshold setting;water types were extracted accurately based on the object-oriented method combined with shape and texture characteristics;construction land and other types were extracted with the maximum likelihood method. The results indicated that the wetland area of Dongping Lake was 179.84 km2 in 2009, with the area of natural wetland 161.32 km2, and artificial wetland 18.52 km2. The overall accuracy of density segmentation and object-oriented method was 88.54%and the overall Kappa coefficient was 0.8776, which were higher than 10.10% and 0.1072 compared with the maximum likelihood method. Wetland information extraction results can provide basic data for Dongping Lake wetland management. Wetland remote sensing information extraction method of density segmentation combined with object-oriented method can provide technical support for rapid and accurate extraction of lake wetland.关键词
密度分割/面向对象/湿地/信息提取/东平湖Key words
Density segmentation/object-oriented/wetland/information extraction/Dongping Lake分类
农业科技引用本文复制引用
王瑷玲,刘洁,王彩艳,王瑞燕..密度分割和面向对象结合的东平湖湿地信息提取[J].山东农业大学学报（自然科学版）,2017,48(1):70-74,5.基金项目
山东省自然科学基金(ZR2013DM006) （ZR2013DM006）
