Integrated pixel-level crack detection and quantification using an ensemble of advanced U-Net architectures

被引:0
|
作者
Rakshitha, R. [1 ]
Srinath, S. [1 ]
Kumar, N. Vinay [2 ]
Rashmi, S. [1 ]
Poornima, B., V [1 ]
机构
[1] JSS Sci & Technol Univ, Dept Comp Sci & Engn, Mysuru, India
[2] Freelance Res, Bangalore, India
关键词
Crack segmentation; Crack quantification; Deep learning; U; -Net; TransUNet; Swin-UNet; Ensemble learning; CONVOLUTIONAL NEURAL-NETWORK; PAVEMENT;
D O I
10.1016/j.rineng.2024.103726
中图分类号
T [工业技术];
学科分类号
08 ;
摘要
Automated pavement crack detection faces significant challenges due to the complex shapes of crack patterns, their similarity to non-crack textures, and varying environmental conditions such as lighting and noise. Traditional methods often struggle to adapt, leading to inconsistent and less accurate results in real-world scenarios. This study introduces a hybrid framework that combines convolutional and transformer-based architectures, leveraging their strengths to achieve reliable crack segmentation and pixel-level quantification. The framework incorporates state-of-the-art deep learning models, including U-Net, Attention U-Net, Residual Attention U-Net (RAUNet), TransUNet, and Swin-Unet. U-Net variants, enhanced with attention mechanisms and residual connections, improve feature extraction and gradient flow, enabling precise delineation of crack boundaries. Transformer-based models like TransUNet and Swin-Unet use self-attention mechanisms to capture both local and global spatial relationships, enhancing robustness across diverse crack patterns. A key contribution of this study is the evaluation of loss functions, including Binary Cross-Entropy (BCE) Loss, Dice Loss, and Binary Focal Loss. Binary Focal Loss proved particularly effective in addressing class imbalance across four benchmark datasets. To further improve segmentation performance, two ensemble strategies were applied: stochastic reordering using logical operations (AND, OR, and averaging) and a weighted average ensemble optimized through grid search. The weighted average ensemble demonstrated superior performance, achieving mean Intersection over Union (mIoU) scores of 0.73, 0.70, 0.78, and 0.86 on the CFD, AgileRN, Crack500, and DeepCrack datasets, respectively. In addition to segmentation, this study developed a method for accurately quantifying crack length and width. By using Euclidean distance along skeletal paths, the algorithm minimized error rates in length and width estimation. This framework provides a scalable and efficient solution for automated pavement crack analysis. It addresses critical challenges in accuracy, adaptability, and reliability under diverse operational conditions, marking significant progress in crack detection technology.
引用
收藏
页数:21
相关论文
共 50 条
  • [41] Pixel-level automatic detection and quantification of running bands on rail surfaces
    Yue, Mingjing
    Yang, Xiancai
    Qian, Yao
    Wang, Ping
    Xu, Jingmang
    Zhang, Allen A.
    AUTOMATION IN CONSTRUCTION, 2024, 165
  • [42] Automatic Pixel-Level Pavement Crack Detection Using Information of Multi-Scale Neighborhoods
    Ai, Dihao
    Jiang, Guiyuan
    Kei, Lam Siew
    Li, Chengwu
    IEEE ACCESS, 2018, 6 : 24452 - 24463
  • [43] Pixel-level pavement crack detection using enhanced high-resolution semantic network
    Xu, Zhengchao
    Sun, Zhaoyun
    Huyan, Ju
    Li, Wei
    Wang, Fengping
    INTERNATIONAL JOURNAL OF PAVEMENT ENGINEERING, 2022, 23 (14) : 4943 - 4957
  • [44] An efficient out-of-distribution pixel-level crack detection framework using prior knowledge
    Li, Hubing
    Gao, Kang
    Liang, Hanbin
    Zhu, Hong
    Yang, Zhiyuan
    Wang, Qiang
    JOURNAL OF BUILDING ENGINEERING, 2024, 94
  • [45] Binocular Video-Based Automatic Pixel-Level Crack Detection and Quantification Using Deep Convolutional Neural Networks for Concrete Structures
    Liu, Liqu
    Shen, Bo
    Huang, Shuchen
    Liu, Runlin
    Liao, Weizhang
    Wang, Bin
    Diao, Shuo
    BUILDINGS, 2025, 15 (02)
  • [46] Improving the Efficiency of Encoder-Decoder Architecture for Pixel-Level Crack Detection
    Chen, Hanshen
    Lin, Huiping
    Yao, Minghai
    IEEE ACCESS, 2019, 7 : 186657 - 186670
  • [47] Optimizing ensemble U-Net architectures for robust coronary vessel segmentation in angiographic images
    Chang, Shih-Sheng
    Lin, Ching-Ting
    Wang, Wei-Chun
    Hsu, Kai-Cheng
    Wu, Ya-Lun
    Liu, Chia-Hao
    Fann, Yang C.
    SCIENTIFIC REPORTS, 2024, 14 (01)
  • [48] A two-stage framework for pixel-level pavement surface crack detection
    Guo, Feng
    Liu, Jian
    Xie, Quanyi
    Yu, Huayang
    ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE, 2024, 133
  • [49] Efficient Road Crack Detection Based on an Adaptive Pixel-Level Segmentation Algorithm
    Safaei, Nima
    Smadi, Omar
    Safaei, Babak
    Masoud, Arezoo
    TRANSPORTATION RESEARCH RECORD, 2021, 2675 (09) : 370 - 381
  • [50] GSD-Net: Compact Network for Pixel-level Graphical Symbol Detection
    Ghosh, Swarnendu
    Shaw, Prasenjit
    Das, Nibaran
    Santosh, K. C.
    2019 INTERNATIONAL CONFERENCE ON DOCUMENT ANALYSIS AND RECOGNITION WORKSHOPS (ICDARW) AND 13TH IAPR INTERNATIONAL WORKSHOP ON GRAPHICS RECOGNITION (GREC 2019), VOL 1, 2019, : 68 - 73