Tropical land use land cover mapping in Pará (Brazil) using discriminative Markov random fields and multi-temporal TerraSAR-X data.

Ron Hagensieker,Ribana Roscher,Johannes Rosentreter,Benjamin Jakimow,Björn Waske

International Journal of Applied Earth Observation and Geoinformation（2017）

引用 20|浏览17

暂无评分

摘要

•Mapping tropical LULC high temporal frequency SAR imagery in Pará (Brazil).•Five class LULC mapping using MRF yields area adjusted accuracies of up to 79%.•Classification of forest class achieves very high accuracy (∼90% UA and PA).•Presentation of a general framework for the mapping of dynamic LULC.

查看译文

关键词

Markov random fields (MRF),Import vector machines (IVM),Multi-temporal LULC mapping,Deforestation,Amazon,SAR

AI 理解论文

溯源树

样例

生成溯源树，研究论文发展脉络

Chat Paper

正在生成论文摘要