题 目：A Bayesian Markov Chain Approach for Land Use Classification Based on Expert Interpretation and Auxiliary Data

主讲人：Professor Chuanrong Zhang

时 间：2016年6月6日（周一）下午15:00

地 点：det365官网网站220学术报告厅

报告内容：Land use maps derived from remotely sensed imagery are usually of insufficient quality for many application purposes. Many conventional classification methods are mainly based on spectral data and cannot classify high level land use classes (e.g., Level III and Level IV classes). Expert knowledge may play an import role in determining image pixel classes through human-computer interactions. The objective of this study is to develop a novel Markov chain geostatistical framework for land use/cover classification with uncertainty assessment based on expert-interpreted pixels from high resolution remotely sensed imagery.

In addition, she will briefly introduce her recent other progress on geostatistic modeling for natural resource and environmental pollutant mapping.

专家简介：Professor Chuanrong Zhang is currently a Professor at University of Connecticut, Storrs. Her research concentrate on GIS, remote sensing, geo-spatial statistics, geocomputation, GIS Cyberinfrastructure, and their applications in land use/cover studies, climate change, managing disasters and natural resources, landscape ecology, environmental planning, as well as transportation studies. So far she has published more than 60 peer-reviewed journal articles (in more than 20 different journals and almost all of them are SCI journals), more than 10 book chapters since 2003. She published her monograph titled “Geospatial Semantic Web” as the first author with Springer in 2015. Her work has been supported from different prestige agents such as National Science Foundation and Department of Energy.