<p>Urban agglomeration is a high-level generalization of urban comprehensive spatial form. Studying its development process is of great significance to the correct planning and healthy management of urban agglomerations. Night light data can detect the weak radiation at night and directly monitor the spatial development pattern of urban agglomerations. At present, the most commonly used data are DMSP/OLS images and NPP/VIIRS images.This paper selected the Chengdu-Chongqing urban agglomeration in China as the study area, and used night-time light images to evaluate its development process from 2000 to 2018. Firstly, in order to obtain long-term stable nighttime light data, we performed calibration and fitting on both datasets. Subsequently, we selected 8 images and used the method of comparing annual statistical data to extract urban built-up areas. Finally, we evaluated the urbanization process. The results showed that: (1) The night light images after normalized processing are more stable and continuous, making the light data under long-term sequences comparable. (2) The results of the statistical data comparison method was numerically verified to show that the error with the statistical data is within ± 5%, and the spatial verification of the results shows that it can basically overlap with the actual location of the city. (3) The “dual-core phenomenon” of the Chengdu-Chongqing urban agglomeration is prominent. In the past 20 years, both Chengdu and Chongqing have expanded rapidly from their original urban centers. The east and south of Chengdu expanded rapidly, and the west and north of Chongqing expanded significantly.</p>