Road cracks are one of the main pavement diseases,and timely and effective detection of road cracks is crucial for pavement maintenance and management.To reduce the model parameter size,improve infer-ence speed,a lightweight road crack detection model based on improved YOLOv8 is proposed.This model embeds a lightweight module called FasterNet into the Neck layer of the YOLOv8 network,that reducing redundant compu-tations and memory access while effectively extracting spatial features.Experimental verification of the algorithm's detection performance is conducted on a self-made road crack dataset.The results show that the improved model significantly reduces model parameters and computational complexity,while also improving Recall and MAP to a certain extent.It ensures the accuracy of road crack detection and facilitates deployment on embedded devices.
NETL
NSTL
万方数据
