Delineation of the Urban Fringe Using Multi-indicators and Deep Neural Network

Abstract

The urban fringe, which located between urban and rural areas, is the most intense area of urban land use change and one of the most likely areas for urban construction land expansion in the future. Accurately and quantitatively identify urban fringe is of great significance for urban planning and sustainable land use. From multi-perspectives of nature, population and socio-economics, this paper attempts to fuse POI big data, remote sensing and population data for delineation of urban fringe. What's more, based on the fusion urban data, we proposed a deep neural network architecture to detect urban fringe, urban and rural area. By using multi- indicators and deep learning, the big city and its surroundings will be divided as urban area, urban fringe, and rural area. Experiments on four cities Guangzhou, Shenzhen, Nanjing and Xi'an have indicated that the proposed method can detect urban fringe effectively, and provide better guidance for urban planning, sustainable development, urban statistical analyses, and overall urban development policy