Human lives and property are threatened by Flash floods (FF) worldwide and as a result of the unprecedented conditions 

of the climate change effects the losses are predicted to increase in the future. As it seems difficult to avoid and prevent 

them, real-time flash flood detections could be an appropriate solution for damage reduction and better management. Cur- 

rently, the development of computer vision applications such as deep learning and AI has been advanced. Although AI 

models have been developed for applications in many fields, their implementations for geosciences are limited based on 

large amounts of training data and the highly required computational infrastructure. Hence, this work aims to train the 

latest YOLOv8 model and apply it to real-time flash flood detection for regions of Korea and possibly for other nations. 

To overcome the shortage of training data, we created small on-site flash flood models and took pictures and footage of 

them. More than 1500 photos of FF were used for model trains and validations gaining a model mean average precision 

of above 60% of all training depths (25, 50, 75, and 100 epochs). Despite some model false positives and missed false 

positive detections using the Korean FF test dataset, the YOLOv8 best model generated bounding boxes (BB) with high 

confidence values in most FF events. Furthermore, the robustness of the model is highlighted by its ability to smoothly 

detect the precise positions of the FF areas with high confidence values (best 0.86) when applied for input footage and 

webcam streams. It is highly encouraged to establish a real-time FF warning system to reduce their negative effects. 

Although YOLO is effective and fast, like other deep learning models, it requires large input data to ensure higher accu- 

racy and confidence. Future works might explore this aspect, particularly the data acquired in light inefficiency to improve 

the model detections at night time.