Vegetation overgrowth in rivers worldwide has been a significant problem due to its potential to reduce flood-flowing capacity and cause biodiversity loss. This study tried to develop a model to predict vegetation recruitment during the initial stage of secondary succession, leading to vegetation overgrowth. This study chose a logistic regression model for predicting vegetation recruitment because of its simplicity and lower computational load than machine learning. The model was designed for the Kinu River in Japan, undergoing significant vegetation overgrowth. Data for the model development was obtained from UAV (unmanned aerial vehicle) aerial surveys and public databases. To ensure the model’s applicability beyond the training rivers, we trained the logistic regression model in its performance across different river flows and geomorphic characteristics, including the normal and flood times and the gravel-bed and sand-bed. The results indicated that the logistic regression model with three explanatory variables — the distance from the river stream, the relative height, and the vegetation existence history —was optimal for all rivers with the F-measures of 0.79 to 0.85. Additionally, using UAV imagery allowed for high spatial resolution in predicting vegetation recruitment. The best model’s prediction map of the vegetation recruitment demonstrated that it could accurately predict vegetation distributions along the main river channel for the gravel and sand beds.