Abstract: Regarding the low efficiency and long cycle of manual inspection of hidden dangers in color steel tiles around railway tracks, this paper designed a method for identifying hidden dangers in color steel tiles along railway lines based on an improved YOLOv8 model. The paper improved the YOLOv8 model by adding Area Loss function to enhance the identifying accuracy, conducted experimental verification on a self-developed dataset for color steel tiles. The results showed that compared with the unimproved YOLOv8 model, the method for identifying hidden dangers in color steel tiles along railway lines based on the improved YOLOv8 model had significant improvement in identifying performance. This method can efficiently and accurately detect hidden dangers in color steel tiles along railway lines, provides technical support for railway safety operation.