YOLOv8-bot: An Improved Road Damage Detection Algorithm Based on YOLOv8

Road damage negatively affects the usage efficiency, reliability, and safety of road transportation, and the detection of road damage has significant economic and social benefits. This study introduces an enhanced road damage detection algorithm for YOLOv8, which integrates the BoTNet network to augment the feature extraction capability of the backbone network, the inclusion of the BoTNet network aims to improve accuracy by capturing long-distance dependencies; optimizing the neck network and introducing a weighted bidirectional feature pyramid BiFPN to enhance the feature fusion ability of the neck network, enhancing the accuracy of road damage detection. The experimental results show that the improved model improves the mAP by 1.2 percentage points and the F1-score by 1.5 percentage points in the road detection task in contrast to the original model.