Rapid identification of urban green space using Planetscope satellite image and artificial intelligence

Abstract

Urban green open space is areas in a city or town filled with vegetation to support socio-ecological functions. These areas have increasingly threatened as a result of being converted to urban infrastructures. As an essential feature of city infrastructure, urban green space should be monitored according to the spatial plan of the city area. However, the space that has been assigned to the urban green space is not a match for its current use. One of the problems that caused urban green space usage mismatch is difficulties in identifying urban green space changes. Planetscope satellite imagery is a high-resolution satellite image that can be used to identify open green spaces in urban areas. In this research, we used an artificial intelligence method to develop a pixel classification process for accurate and efficient identification of the green open space. The results showed that Planetscope satellite imagery and artificial intelligence methods had 99% accuracy in monitoring green open spaces. The use of this technology can assist in the early detection of green open space changes effectively and efficiently.