Multi-scale feature fusion network for pixel-level pavement distress detection

Automatic pavement distress detection is essential to monitoring and maintaining pavement condition. Currently, many deep learning-based methods have been utilized in pavement distress detection. However, distress segmentation remains as a challenge under complex pavement conditions. In this paper,...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Automation in construction Jg. 141; S. 104436
Hauptverfasser: Zhong, Jingtao, Zhu, Junqing, Huyan, Ju, Ma, Tao, Zhang, Weiguang
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier B.V 01.09.2022
Schlagworte:
Deep learning
Encoder-decoder architecture
Feature fusion
Pavement distress
Semantic segmentation
Unmanned aerial vehicle (UAV)
Deep learning
Unmanned aerial vehicle (UAV)
Pavement distress
Feature fusion
Encoder-decoder architecture
Semantic segmentation
ISSN:0926-5805, 1872-7891
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Automatic pavement distress detection is essential to monitoring and maintaining pavement condition. Currently, many deep learning-based methods have been utilized in pavement distress detection. However, distress segmentation remains as a challenge under complex pavement conditions. In this paper, a novel deep neural network architecture, W-segnet, based on multi-scale feature fusions, is proposed for pixel-wise distress segmentation. The proposed W-segnet concatenates distress location information with distress classification features in two symmetric encoder-decoder structures. Three major types of distresses: crack, pothole, and patch are segmented and the results were discussed. Experimental results show that the proposed W-segnet is robust in various scenarios, achieving a mean pixel accuracy (MPA) of 87.52% and a mean intersection over union (MIoU) of 75.88%. The results demonstrate that W-segnet outperforms other state-of-the-art semantic segmentation models of U-net, SegNet, and PSPNet. Comparison of cost of model training and inference indicates that W-segnet has the largest number of parameters, which needs a slightly longer training time while it does not increase the inference cost. Four public datasets were used to test the generalization ability of the proposed model and the results demonstrate that the W-segnet possesses well segmentation performance. •W-segnet with two encoder-decoder architectures is proposed for multi type of pavement distresses segmentation.•A challenging benchmark dataset was established via UAV for distress segmentation.•W-segnet is applied in segmenting pavement crack, pothole, and patch images under various scenarios.•W-segnet outperforms U-net, SegNet, and PSPNet in self-built and open-source dataset with higher accuracy.•W-segnet possesses a higher inference speed for model deployment in practical application.
ISSN:0926-5805
1872-7891
DOI:10.1016/j.autcon.2022.104436