The Crack Diffusion Model: An Innovative Diffusion-Based Method for Pavement Crack Detection

被引:6
|
作者
Zhang, Haoyuan [1 ]
Chen, Ning [1 ]
Li, Mei [1 ]
Mao, Shanjun [1 ]
机构
[1] Peking Univ, Inst Remote Sensing & Geog Informat Syst, Beijing 100871, Peoples R China
基金
国家重点研发计划;
关键词
pavement crack detection; diffusion model; multi-task; spatial continuity;
D O I
10.3390/rs16060986
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
Pavement crack detection is of significant importance in ensuring road safety and smooth traffic flow. However, pavement cracks come in various shapes and forms which exhibit spatial continuity, and algorithms need to adapt to different types of cracks while preserving their continuity. To address these challenges, an innovative crack detection framework, CrackDiff, based on the generative diffusion model, is proposed. It leverages the learning capabilities of the generative diffusion model for the data distribution and latent spatial relationships of cracks across different sample timesteps and generates more accurate and continuous crack segmentation results. CrackDiff uses crack images as guidance for the diffusion model and employs a multi-task UNet architecture to predict mask and noise simultaneously at each sampling step, enhancing the robustness of generations. Compared to other models, CrackDiff generates more accurate and stable results. Through experiments on the Crack500 and DeepCrack pavement datasets, CrackDiff achieves the best performance (F1 = 0.818 and mIoU = 0.841 on Crack500, and F1 = 0.841 and mIoU = 0.862 on DeepCrack).
引用
收藏
页数:19
相关论文
共 50 条
  • [1] A novel blade crack detection method based on diffusion model with acoustic-vibration fusion
    Zhao, Xun
    Xu, Feiyun
    Song, Di
    Shen, Junxian
    Ma, Tianchi
    2023 IEEE 21ST INTERNATIONAL CONFERENCE ON INDUSTRIAL INFORMATICS, INDIN, 2023,
  • [2] Pavement Crack Detection Method Based on Deep Learning Models
    Hu, Guo X.
    Hu, Bao L.
    Yang, Zhong
    Huang, Li
    Li, Ping
    WIRELESS COMMUNICATIONS & MOBILE COMPUTING, 2021, 2021
  • [3] DIFFUSION-BASED MODEL OF PHOTORECEPTION
    YASUI, S
    ANDRESEN, MC
    BROWN, AM
    BIOPHYSICAL JOURNAL, 1978, 21 (03) : A117 - A117
  • [4] Crack-Photolithography for Membrane-Free Diffusion-Based Micro/Nanofluidic Devices
    Kim, Minseok
    Kim, Taesung
    ANALYTICAL CHEMISTRY, 2015, 87 (22) : 11215 - 11223
  • [5] Pavement Crack Detection and Segmentation Method Based on Improved Deep Learning Fusion Model
    Feng, Xiaoran
    Xiao, Liyang
    Li, Wei
    Pei, Lili
    Sun, Zhaoyun
    Ma, Zhidan
    Shen, Hao
    Ju, Huyan
    MATHEMATICAL PROBLEMS IN ENGINEERING, 2020, 2020
  • [6] ITERATIVE DIFFUSION-BASED ANOMALY DETECTION
    Mishne, Gal
    Cohen, Israel
    2017 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP), 2017, : 1682 - 1686
  • [7] Diffusion-based kernel matrix model for face liveness detection
    Yu, Changyong
    Yao, Chengtang
    Pei, Mingtao
    Jia, Yunde
    IMAGE AND VISION COMPUTING, 2019, 89 : 88 - 94
  • [8] A Diffusion-based Method for Entity Search
    Torres-Tramon, Pablo
    Timilsina, Mohan
    Hayes, Conor
    2019 13TH IEEE INTERNATIONAL CONFERENCE ON SEMANTIC COMPUTING (ICSC), 2019, : 16 - 23
  • [9] A method for superfine pavement crack continuity detection based on topological loss
    Jia, Guohui
    Lu, Yang
    Song, Weidong
    He, Huan
    Ma, Rongchen
    ELECTRONICS LETTERS, 2023, 59 (19)
  • [10] Pavement Crack Detection and Quantification Based on Scanning Grid and Projection Method
    Sun, Zhaoyun
    Pei, Lili
    Yuan, Bo
    Du, Yaohui
    Li, Wei
    Han, Yuxi
    PROCEEDINGS OF THE 8TH INTERNATIONAL CONFERENCE ON CIVIL ENGINEERING, 2022, : 273 - 281