Abstract: The new generation of geostationary weather satellite imagers has the advantage of high spatial and temporal resolution, and cloud observations from the imagers are widely used in various studies in the field of meteorology. Due to the observation mode of the satellite and the curvature of the Earth, it has a parallax problem, the impact of which needs to be considered when it is jointly applied with other information. To address the parallax problem of the Fengyun-4A imager, the Advanced Geosynchronous Radiation Imager (AGRI), a sensitivity analysis was first performed using the simulated cloud top height and the actual satellite zenith angle, and the results confirmed that the higher the cloud top height or the larger the satellite zenith angle, the greater the parallax, especially for the imager with higher spatial resolution. Cloud data from the Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observation (CALIPSO) and three sets of typhoon best track sets were then used to examine the accuracy of AGRI cloud mask products and typhoon center positioning before and after the parallax correction. The results demonstrate the effectiveness of the parallax correction method used, and point out that parallax cannot be neglected in the accurate quantitative application of geostationary meteorological satellite datasets.