Beyond surveys: A High-Precision Wealth Inequality Mapping of China's Rural Households Derived from Satellite and Street View Imageries

Abstract

Wide coverage and high-precision rural household wealth data is an important support for the effective connection between the national macro rural revitalization policy and micro rural entities, which helps to achieve precise allocation of national resources. However, due to the large number and wide distribution of rural areas, wealth data is difficult to collect and scarce in quantity. Therefore, this article attempts to integrate "sky" remote sensing images with "ground" village street view imageries to construct a fine-grained "computable" technical route for rural household wealth. With the intelligent interpretation of rural houses as the core, the relevant wealth elements of image data were extracted and identified, and regressed with the household wealth indicators of the benchmark questionnaire to form a high-precision township scale wealth prediction model (r=0.85); Furthermore, a national and township scale map of rural household wealth in China was promoted and drawn. Based on this, this article finds that there is a "bimodal" pattern in the distribution of wealth among rural households in China, which is reflected in a polarization feature of "high in the south and low in the north, and high in the east and low in the west" in space. This technological route may provide alternative solutions with wider spatial coverage and higher accuracy for high-cost manual surveys, promote the identification of shortcomings in rural construction, and promote the precise implementation of rural policies.