Abstract: Landslide activity, an indicator for characterizing the kinematic state of a landslide mass, is a crucial component of landslide risk assessment. However, conventional methods for landslide activity assessment suffer from limitations such as high dependence on subjective judgment, low efficiency, and high uncertainty. To overcome these limitations, this study proposed an innovative assessment method based on integrated remote sensing and an ordered logistic regression model. Specifically, based on multi-source remote sensing data and related survey results, landslide hazard identification and assessment metric extraction were achieved by integrating interferometric synthetic aperture radar (InSAR)-based surface deformation monitoring, optical interpretation of hazards, and spatial analysis. Active landslide samples were then established through field surveys. Finally, an assessment model for landslide activity rating was constructed based on ordered logistic regression. The proposed method proved effective in assessing landslide activity in the lower reaches of the Jinsha River in Yunnan Province, yielding an assessment accuracy of 75.58%. The analysis of application results reveals that the assessment metrics significantly affecting landslide activity in the study area included the ratio of the deformation area to the total study area, the average deformation rate, and the angle between the deformation slope and direction. The proposed method offers distinct advantages in terms of efficiency, accuracy, and objectivity, holding considerable application value for the investigation and assessment of regional landslide risks.