MADF-YOLOv8: A Lightweight Model for Road Distress Detection Based on Adaptive Multiscale Feature Fusion

Tao OuYang; Haohui Yu; Guanlin Pan; Yan Cui; Qingling Chang; Xiulong Fu

doi:10.26689/jera.v9i6.13159

Tao OuYang Wuyi University, Jiangmen 529020, Guangdong, China
Haohui Yu Wuyi University, Jiangmen 529020, Guangdong, China
Guanlin Pan Wuyi University, Jiangmen 529020, Guangdong, China
Yan Cui Wuyi University, Jiangmen 529020, Guangdong, China
Qingling Chang Wuyi University, Jiangmen 529020, Guangdong, China
Xiulong Fu Wuyi University, Jiangmen 529020, Guangdong, China

DOI: https://doi.org/10.26689/jera.v9i6.13159

Keywords: Road distress detection, Multi-scale feature fusion, YOLOv8

Abstract

Efficient road distress detection is crucial for transportation safety. To address the challenge of balancing detection accuracy, efficiency, and multi-scale feature fusion in existing methods, this paper proposes a lightweight model named MADF-YOLOv8. The model enhances multi-scale feature extraction capability by introducing the Multi-Scale Ghost Residual Convolution (MSGRConv) and the Multiscale Adaptive Feature Processing Module (MAFP). Furthermore, it constructs a Multi-scale Dynamic sampling Bidirectional Feature Pyramid Network (MD-BiFPN) and incorporates the C2f-Faster module to optimize feature fusion efficiency. Experiments on the RDD2022 dataset demonstrate that the proposed model achieves a mean Average Precision at 0.5 Intersection over Union (mAP@0.5) of 88.6% with only 2.312 million parameters. Its overall performance surpasses various mainstream detectors, achieving an exceptional balance between accuracy and efficiency.

References

Liu W, Lu H, Fu H, et al., 2023, Learning to Upsample by Learning to Sample. Proceedings of the IEEE/CVF International Conference on Computer Vision, 2023.

Chen J, Kao S, He H, et al., 2023, Run, Don’t Walk: Chasing Higher FLOPS for Faster Neural Networks. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023.

Arya D, Maeda H, Ghosh S, et al., 2024, RDD2022: A Multi-National Image Dataset for Automatic Road Damage Detection. ArXiv. https://doi.org/10.48550/arXiv.2209.08538

Glenn J, Alex S, Jirka B, 2023, Ultralytics Yolov8, 8.

Wang C, Bochkovskiy A, Liao H, YOLOv7: Trainable Bag-of-Freebies Sets New State-of-the-Art for Real-Time Object Detectors. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023.

Wang C, Yeh I, Mark L, 2024, Yolov9: Learning What You Want to Learn Using Programmable Gradient Information. Proceedings of the European Conference on Computer Vision, 2024.

Wang A, Chen H, Liu L, et al., 2024, Yolov10: Real-Time End-to-End Object Detection, 37(10): 7984–8011.

Khanam R, Hussain M, 2024, Yolov11: An Overview of the Key Architectural Enhancements.

Naddaf S, Naddaf S, Kashai A, et al., 2020, An Efficient and Scalable Deep Learning Approach for Road Damage Detection. Proceedings of the 2020 IEEE International Conference on Big Data

Wang X, Gao H, Jia Z, et al., 2023, BL-YOLOv8: An Improved Road Defect Detection Model based on YOLOv8, 23(20): 8361.

Ren S, He K, Girshick R, et al., 2016, Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks, 39(6): 37–49.