首页 | 本学科首页   官方微博 | 高级检索  
     

基于对象级分类的土地覆盖动态变化及趋势分析
引用本文:吴田军,胡晓东,夏列钢,骆剑承,沈占锋,吴炜.基于对象级分类的土地覆盖动态变化及趋势分析[J].遥感技术与应用,2014,29(4):600-606.
作者姓名:吴田军  胡晓东  夏列钢  骆剑承  沈占锋  吴炜
作者单位:(1.中国科学院遥感与数字地球研究所,北京100101;2.中国科学院大学,北京100049)
基金项目:国家自然科学基金项目(41101398;41271367);国家863计划项目(2013AA12A401);高分专项(民用部分)科研项目(03-Y30B06-9001-13/15);中国科学院重点部署项目课题(KZZD-EW-07-02);国家科技支撑计划重点项目课题(2011BAH06B02);国家科技支撑计划课题(2012BAH33B01);国家国际科技合作计划项目(2010DFA92720)
摘    要:以广东省东莞市2005~2008年SPOT 5遥感影像为主要数据源,采用对象级分类后比较的变化检测技术,从土地覆盖类型的面积总量、相互转移等方面多层次分析了研究区域4 a土地覆盖的变化情况。在此基础上,利用马尔科夫链模型对该区域未来5 a的土地覆盖动态变化及演变趋势进行了分析和预测,为广东省土地覆盖变化研究提供典型案例分析,以达到全面把握研究区土地覆盖变化规律的目标。结果表明:研究区域的土地覆盖变化基本趋势表现为城镇建设用地总量持续增加以及耕地和园林地面积总量减少,其中城镇建设用地和农业耕地变化幅度较大,其他类型土地变化相对稳定。增加的城镇建设用地主要来源于耕地的人为减少,农业垦殖环境将趋于恶化。研究结果可为研究区合理有效地利用土地以促进土地可持续发展、推进城市化进程奠定决策上的技术基础。

关 键 词:土地覆盖  变化检测  对象级分类  趋势预测  马尔科夫链  

Analysis on Dynamic Change and Tendency of Land-cover based on Object\|oriented Classification
Wu Tianjun,Hu Xiaodong,Xia Liegang,Luo Jiancheng,Shen Zhanfeng,Wu Wei.Analysis on Dynamic Change and Tendency of Land-cover based on Object\|oriented Classification[J].Remote Sensing Technology and Application,2014,29(4):600-606.
Authors:Wu Tianjun  Hu Xiaodong  Xia Liegang  Luo Jiancheng  Shen Zhanfeng  Wu Wei
Affiliation:( 1.Institute of Remote Sensing and Digital Earth,Chinese Academy of Science,Beijing,100101,China;; 2.University of Chinese Academy of Sciences,Beijing 100049,China)
Abstract:In order to accurately analyze the dynamic change of land cover,this paper presented a change detection method based on object\|oriented classification.For two images with high spatial resolutions,the proposed approach was compared the land cover type after they are classified by object\|oriented method.According to the SPOT5 remote sensing imagines of Dongguan city,Guangdong Province between the years of 2005 and 2008,this paper adopted the change detection technique based on object\|oriented classification and further analyzed quantitatively the land cover change in the aspects of area and reciprocal transformation.The object\|oriented classification\|based change detection method is certified more accuracy than pixel\|oriented classification\|based change detection method.Moreover,using Markov chain model,the research simulated and predicted the dynamic change and evolution tendency of land cover over the study area in the next five years.Through this process of study,we provided a typical case for grasping the rules of land cover dynamic change over the study area.The results showed that construction land are still increasing,whereas cultivated land and wood land are decreasing.Furthermore,the change area of construction land and cultivated land were larger than other classes of land cover.In addition,the increasing of construction land is mainly caused by the decreasing of cultivated land,which speeds up the deterioration of the cultivated environment in the study area.In conclusion,the research could provide technical support for effectively using land and promoting the sustainable development of land and urbanization.
Keywords:Land cover  Change detection  Object-oriented classification  Forecast on trend  Markov chain  
本文献已被 CNKI 等数据库收录!
点击此处可从《遥感技术与应用》浏览原始摘要信息
点击此处可从《遥感技术与应用》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号