Application of remote sensing data in digital transformation of land resource management in Kim Thanh commune, Hai Phong city

Abstract

In the context of digital transformation in public administration, especially in land resource management, remote sensing technology offers timely, objective and scalable solutions. This study investigates the integration of Sentinel-2 satellite imagery and object-based classification techniques to support digital land management in Kim Thanh commune, a newly established administrative unit in Hai Phong city, Vietnam. Sentinel-2 imagery acquired on 20 March 2025 was processed, atmospherically corrected and segmented using the SNIC algorithm. Land cover was classified into three main types: water body, agricultural land and non-agricultural land using the Random Forest algorithm on the Google Earth Engine platform. The classification achieved a high overall accuracy of 93.17% and a Kappa coefficient of 0.895, with User Accuracy ranging from 87.40% to 100%. The spatial distribution of land use types suggests reasonable zoning patterns that can be directly integrated into digital cadastral systems and spatial planning tools. The results of the study demonstrate the potential of combining remote sensing and cloud computing technologies to improve transparency, efficiency and decision-making in land governance, contributing to the broader goals of the national digital transformation strategy in Vietnam.