Lane detection is a critical component of autonomous driving systems, enabling vehicles to identify and navigate within lanes accurately. This paper presents a novel approach to enhancing lane detection ac curacy using the Mask R-CNN algorithm. By leveraging the capabilities of Mask R-CNN, the proposed algorithm demonstrates efficient and precise detection of road lanes, including the classification of lane types and angle evaluation for steer ing purposes. The algorithm’s functionality encompasses determining bounding boxes of lanes, angle evaluation through image cropping, classification, and lane data configuration for schematic environ mental surveillance. Through extensive testing, the algorithm has shown superior performance in scenarios with challenging conditions such as insufficient lighting and lane line degradation. The results indicate a significant improvement in lane detection accuracy, making it a promising solution for advancing the capabilities of autonomous driving systems.