Pilot development of a satellite image-based spatial analysis tool to support rural spatial planning

This study is about the pilot development of a satellite image-based spatial analysis tool to support rural spatial planning. For the sustainable development and systematic use of space in rural areas, accurate, integrated, and data-driven decision-making is essential. However, the dispersed management of rural-related data, non-standardized formats, and the absence of periodic monitoring systems are acting as obstacles to establishing effective plans. To address this issue, this study applied the following methodology. First, rural-related data stored separately by the central government and local governments were collected and processed into standardized spatial data based on Geographic Information System (GIS).  Second, a satellite image-based facility detection and classification tool was developed for periodic and efficient monitoring of rural facilities. The target facilities were selected as livestock, factories, and solar panels, and the latest deep learning model based on HRNet-OCR architecture was implemented and optimized for the rural environment. For the training and validation data, the mosaic image of the Korean Peninsula (2019~2020) produced by KOMPSAT satellite images was used, which provided a high-resolution spatial resolution of 1m and multiple spectral bands to enable the analysis of various indicator characteristics. Third, to verify the effectiveness of the developed tool, Seosan City, Anseong City, Naju City, and Geochang County in the Republic of Korea were selected as pilot areas. These regions were deemed to represent diverse rural characteristics and facility distributions. Finally, a user-friendly web-based information support tool was developed by integrating processed rural data and satellite image analysis results. The results of this study are expected to be utilized as foundational data for establishing rural spatial plans to support rural spatial restructuring and regeneration, and the developed spatial analysis tool is deemed capable of contributing to the formulation of more efficient and sustainable rural development strategies by providing a data-driven decision support system to rural policymakers.