Forest resource is often changed by the influence of human activities and natural damages. It is important to detect those change in time for forestry management planning. In China, the traditional way to obtain the c...Forest resource is often changed by the influence of human activities and natural damages. It is important to detect those change in time for forestry management planning. In China, the traditional way to obtain the change information of forest resource is carried out by means of "forest resource inventory and planning". There are several disadvantage within this means, for example, the period of re inventory is longer, the quality of sub compartment distribution map is poor, etc.. So we tried to find a good way to detect the forest resource change and to get finely result in order to update the forest data in time. In this paper, we introduced the study on detecting the forest resource change information and forest mapping using remote sensing and GIS techniques combined the conventionality way. First, the forest change area were identified from TM image and created a new sub compartment boundary map, then updating the old sub compartment boundary map using this new map. Second, updating the attribution database according to the sub compartment level field inventory data and the new sub compartment boundary map. Finally some comparison were made between the result of monitoring and the local field data, the accuracy is good. The study results showed this paper can be used as a working plan for further forest GIS development, where remote sensing is also utilized as a part of forest resource information management system.展开更多
该文使用近10 a 4个时相的江苏全省Landsat遥感影像,在数据预处理的基础上提取归一化植被指数、比值植被指数、土壤调节植被指数、增强型植被指数、大气阻抗植被指数等5种植被指数,并进行主成分分析。运用最大似然法、随机森林法和光谱...该文使用近10 a 4个时相的江苏全省Landsat遥感影像,在数据预处理的基础上提取归一化植被指数、比值植被指数、土壤调节植被指数、增强型植被指数、大气阻抗植被指数等5种植被指数,并进行主成分分析。运用最大似然法、随机森林法和光谱角填图法进行分类,结合小班数据,对各方法的分类结果进行精度评价。评价结果表明,光谱角填图分类法在杨树信息提取时精度更高,对杨树的区分精度也达到42.67%。展开更多
文摘Forest resource is often changed by the influence of human activities and natural damages. It is important to detect those change in time for forestry management planning. In China, the traditional way to obtain the change information of forest resource is carried out by means of "forest resource inventory and planning". There are several disadvantage within this means, for example, the period of re inventory is longer, the quality of sub compartment distribution map is poor, etc.. So we tried to find a good way to detect the forest resource change and to get finely result in order to update the forest data in time. In this paper, we introduced the study on detecting the forest resource change information and forest mapping using remote sensing and GIS techniques combined the conventionality way. First, the forest change area were identified from TM image and created a new sub compartment boundary map, then updating the old sub compartment boundary map using this new map. Second, updating the attribution database according to the sub compartment level field inventory data and the new sub compartment boundary map. Finally some comparison were made between the result of monitoring and the local field data, the accuracy is good. The study results showed this paper can be used as a working plan for further forest GIS development, where remote sensing is also utilized as a part of forest resource information management system.
文摘该文使用近10 a 4个时相的江苏全省Landsat遥感影像,在数据预处理的基础上提取归一化植被指数、比值植被指数、土壤调节植被指数、增强型植被指数、大气阻抗植被指数等5种植被指数,并进行主成分分析。运用最大似然法、随机森林法和光谱角填图法进行分类,结合小班数据,对各方法的分类结果进行精度评价。评价结果表明,光谱角填图分类法在杨树信息提取时精度更高,对杨树的区分精度也达到42.67%。
