Objectives China has formed three major economic belts according to regional resource condition, economic foundation, and development orientation, namely the eastern, central, and western economic belts. A comprehensive analysis of economic development across China's three major economic belts is essential. It helps reveal regional disparities, supports policy evaluation and formulation, and contributes to improvements in living standards. Traditional regional economic research mainly relies on socio-economic statistical data. Although effective, such data lack spatial continuity and fine-scale spatial representation. Remote sensing imagery can extract data and spatial distribution information on land use, urban construction, and development activity from a macroscopic perspective, playing an important role in regional economic research. However, current research still primarily uses the single data source, such as night-time light data, for economic development studies. The integration of multi-source remote sensing data with advanced feature extraction methods provides new opportunities for comprehensive regional economic assessment.

Methods In response to the current lack of studies utilizing multi-source remote sensing data for regional economic analysis, a comprehensive evaluation and analysis method is proposed. This method employs Sentinel-2 data, NPP-VIIRS night-time light data, as well as population and gross domestic product (GDP) data, extracting features that reflect regional economic development from different perspectives. A vision transformer (ViT)-based semantic segmentation model is applied to Sentinel-2 imagery to generate detailed land use classification maps. The threshold method is also applied to extract built-up areas from night-time light data. Nine evaluation indicators are constructed by combining the results of multi-source remote sensing image extraction with GDP and population data. By constructing a regional economic scoring system based on the analytic hierarchy process, the method provides a comprehensive analysis of the economic development status of representative regions within economic belts. This approach offers strong data support for understanding the current state and disparities among the three major economic belts.

Results The results indicate that the regional economic scores for Jiangsu Province are 0.660 5, 0.706 1, and 0.767 1 across the three study periods, followed by Hubei Province with scores of 0.519 9, 0.530 7, and 0.573 4, and the Ningxia Hui Autonomous Region with scores of 0.294 3, 0.321 2, and 0.360 1, respectively. All three regions show consistent upward trends, reflecting overall improvement in regional development levels. Jiangsu Province maintains the highest score throughout the study period, showing strong economic vitality and a solid development foundation. Hubei Province demonstrates steady growth and structural stability, although a temporary slowdown is observed from 2020 to 2021, corresponding to the impact of the pandemic. However, subsequent data also show that Hubei Province recovers relatively quickly after the pandemic. In contrast, the Ningxia Hui Autonomous Region remains at a comparatively lower development level. Nevertheless, its economic score increases steadily, and its average growth rate ranks highest among the three regions, indicating strengthening development momentum within the western economic belt despite persistent structural disparities.

Conclusions The integration of multi-source remote sensing data and socio-economic indicators enhances the accuracy and spatial explicitness of regional economic evaluation. The combination of ViT-based land use classification, night-time light analysis, and a regional economic scoring system enables multi-temporal assessment of development disparities among economic belts. The results reflect structural differences consistent with regional development patterns and recent policy influences. This framework provides a scalable and data-driven reference for monitoring regional economic dynamics and supporting coordinated development planning.