The improved model of YOLOV11 integrating Starnet
DOI:
https://doi.org/10.63313/JCSFT.9020Keywords:
Road crack detection, RDD2022 dataset, Yolov11, C3k2-star backboneAbstract
Due to the small size, widespread distribution, and complex background of road cracks, traditional detection methods are prone to misdetection and missed detection. To improve the model's detection performance, we propose an im-proved YOLOv11 model for road crack detection. The model uses Starnet as the backbone network and dynamically adjusts its large spatial receptive field to more accurately capture the details of road cracks. Validation on the RDD2022 dataset shows that the improved model achieves an average precision (mAP) of 0.586, which is a 4.8 percentage point improvement over the traditional model. These improvements significantly enhance detection accuracy and efficiency, providing a more reliable and efficient solution for road crack detection.
References
[1] Zhou C, Yang Y, Lin L, et al. Research on pavement crack detection technology based on improved YOLO11[C]//2025 40th Youth Academic Annual Conference of Chinese Associa-tion of Automation (YAC). IEEE, 2025: 2220-2223.
[2] Meng A, Zhang X, Yu X, et al. Investigation on lightweight identification method for pave-ment cracks[J]. Construction and Building Materials, 2024, 447: 138017.
[3] Hu G X, Hu B L, Yang Z, et al. Pavement crack detection method based on deep learning models[J]. Wireless Communications and Mobile Computing, 2021, 2021(1): 5573590.
[4] Hong P T H, Hang T T T, Van Giap D, et al. EB-UNet++: An enhanced crack segmentation network combining EfficientNet-B2 and UNet++ with boundary extraction module[J]. Journal of Military Science and Technology, 2025 (IITE): 148-159.
[5] Sapkota R, Karkee M. Ultralytics YOLO Evolution: An Overview of YOLO26, YOLO11, YOLOv8 and YOLOv5 Object Detectors for Computer Vision and Pattern Recognition[J]. arXiv preprint arXiv:2510.09653, 2025.
[6] Wu D. YOLO-Based Enhanced Crack Detection Algorithm for Underground Pipeline Sce-narios[J]. Traitement du Signal, 2025, 42(2): 787.
[7] Żarski M, Wójcik B, Miszczak J A. KrakN: Transfer learning framework and dataset for in-frastructure thin crack detection[J]. SoftwareX, 2021, 16: 100893.
[8] YOLOv11-TDSP: Lightweight and high-precision abnormal tooth detection model for oral panoramic films[J]. Journal of Southern Medical University, 2025, 45(8): 1791.
Downloads
Published
Issue
Section
License
Copyright (c) 2025 by author(s) and Erytis Publishing Limited.
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
How to Cite
