Japan Geoscience Union Meeting 2016

Presentation information

International Session (Poster)

Symbol H (Human Geosciences) » H-TT Technology & Techniques

[H-TT09] Geographic Information Systems and Cartography

Sun. May 22, 2016 5:15 PM - 6:30 PM Poster Hall (International Exhibition Hall HALL6)

Convener:*Takashi Oguchi(Center for Spatial Information Science, The University of Tokyo), Yuji Murayama(Graduate School of Life and Environmental Sciences), Masatoshi Arikawa(Center for Spatial Information Science, The University of Tokyo)

5:15 PM - 6:30 PM

[HTT09-P07] Modeling Urban Land Use / Cover Changes Based on Machine Learning Techniques: A Case study of Shanghai, China

*Hao GONG1, Yuji Murayama1 (1.Graduate Schools and Programs Life and Environmental Sciences, University of Tsukuba)

Keywords:LULC, Machine learning, Shanghai, Urban growth modeling

Urban growth is one of the most important topics in urban studies. A city is considered as a complex system. It consists of numerous interactive sub-systems and is affected by various factors including governmental land policies, population growth, transportation infrastructure and market behavior. To understand the driving forces of the urban form and structural changes, the satellite-based estimation is considered as the appropriate methods to monitor these dynamic changes in a long term.
Based on previous studies, classified Landsat satellite images are used to monitor the temporal changes of land use and land cover (LULC) for the study area. Furthermore, modeling and simulation are believed to be powerful tools to explore the mechanisms of urban evolution and to support the planning in growth management. In this study, authors use the social and geographical factors to model and simulate the urban growth in Shanghai. Finally, an attempt is made to utilize two machine learning models (the deep convolutional network and multi-layer perceptron neural network) to predict the future changes in the land use / cover, and compare the performance of two models.