Mapping urban green space of the 36 major cities in China using ZY-3 satellite images based on a feature robust to low-brightness pixels

Abstract

Urban green spaces (UGS), integral to modern urban life, play a pivotal role in shaping ecological, low-carbon, resilient, and livable cities. Understanding the microstructure and macro patterns of UGS is necessary but limited. In this context, we investigated the spatial patterns in the 36 major cities in China using Ziyuan-3 remote sensing images. We used a method for extracting UGS suitable for complex urban environments, and then selected 5 landscape pattern indices such as percent of landscape and patch density to analyse the spatial pattern of UGS in the 36 cities. The results show that: (1) The overall accuracy, recall, F1 value, and intersection over union of the extraction results of UGS were 94.62%, 94.11%, 94.36%, and 89.33% respectively; (2) Overall, the green space rate of the 36 cities exhibited a spatial distribution pattern of “high in the east and low in the west, low in the north and high in the south”; (3) Variations among cities were weak in terms of patch shape, but significant in patch density. The findings of this study cater to the escalating demands of urban planning and management.