论文部分内容阅读
分别利用1995—2000年的ERS-1/2串行数据和2005年的Envisat ASAR数据对我国东北林区进行森林制图研究。针对ERS-1/2数据相干模型,采用一种不依靠地面实况数据而是基于MODIS植被连续覆盖产品进行训练的方法,从而实现进行大区域森林蓄积量分级制图的目的。分级制图包括0～20、>20～50、>50～80和>80m3/hm24个蓄积量等级。基于Envisat ASAR数据,采用面向对象的分类方法,进行自动化森林和非森林分类处理。基于2005年Landsat TM-5分类结果的交叉验证表明:这2种传感器SAR数据均可用于大区域森林制图。2期森林制图结果为进一步的森林变化分析以及制图更新研究提供支持。
The forest mapping of the forest region in Northeast China was studied using the ERS-1/2 serial data from 1995 to 2000 and the Envisat ASAR data from 2005 respectively. For the ERS-1/2 data coherence model, a method of continuously covering products based on MODIS vegetation without ground-truth data is adopted, so as to achieve the purpose of hierarchical mapping of large-scale forest volume. Grading chart includes 0 ~ 20,> 20 ~ 50,> 50 ~ 80 and> 80m3 / hm24 volume accumulation level. Based on Envisat ASAR data, an object-oriented classification method is used for automated forest and non-forest classification. Cross-validation based on the 2005 Landsat TM-5 classification results showed that both of these sensor SAR data were available for large-area forest mapping. Phase 2 forest mapping provides support for further forest change analysis and mapping update studies.