To solve the problem of limited and difficult data collection in traditional road crack image segmentation, a Convolutional Block Implicit Diffusion Model (CBIDM) based on convolutional block attention mechanism is proposed to generate and enhance crack images, highlighting features with high correlation with cracks in the image, making the model more sensitive to the connection between small cracks and coarse and fine cracks. Based on the public dataset CRACK500, experiments were conducted on the model proposed in this paper. The results showed that after expanding the original data with generated images in a 1:1 ratio, the U-Net segmentation model was trained and tested. The mIoU and mAP indicators for crack segmentation were improved by 2.63% and 4.84% respectively compared to the original dataset, with an average accuracy of 97.05%. This verified that using the proposed model for data generation and enhancement can effectively improve the performance of crack image segmentation.

Zhang Pengwei, Zhao Chen, Chen Jingxia, Wang Zikai. Crack Image Augmentation and Segmentation Based on Convolutional Block Attention Implicit Diffusion Model. Academic Journal of Computing & Information Science (2024), Vol. 7, Issue 7: 31-40. https://doi.org/10.25236/AJCIS.2024.070705.

[1] Ministry of Transport of the People's Republic of China The 14th Five Year Plan for the Development of Modern Comprehensive Transportation System [R] Railway Technical Supervision, 2022, 50 (2): 9-23, 27.

[2] Kheradmandi N, Mehranfar V. A critical review and comparative study on image segmentation-based techniques for pavement crack detection [J]. Construction and Building Materials, 2022, 321: 126162.

[3] Song Zegang, Liu Yanli, Zhang Changxing. Application and Development Review of Bridge Crack Detection Based on Machine Vision [J]. Science and Technology and Engineering, 2023, 23 (30): 12796-12805.

[4] Iyer S, Sinha S K. A robust approach for automatic detection and segmentation of cracks in underground pipeline images [J]. Image and Vision Computing, 2005, 23(10): 921-933.

[5] Yiyang Z. The design of glass crack detection system based on image preprocessing technology[C]// Proc of the 7th joint international information technology and artificial intelligence conference. Piscataway, NJ: IEEE Press, 2014: 39-42.

[6] Hoang N D, Huynh T C, Tran X L, et al. A novel approach for detection of pavement crack and sealed crack using image processing and salp swarm algorithm optimized machine learning[J]. Advances in Civil Engineering, 2022, 2022. 

[7] Akagic A, Buza E, Omanovic S, et al. Pavement crack detection using Otsu thresholding for image segmentation[C]//Proc of the 41st international convention on information and communication technology, electronics and microelectronics (MIPRO). Piscataway, NJ: IEEE, 2018: 1092-1097.

[8] Hertz J A. Introduction to the theory of neural computation [M]. Crc Press, 2018.

[9] Cheng H D, Shi X J, Glazier C. Real-time image thresholding based on sample space reduction and interpolation approach [J]. Journal of computing in civil engineering, 2003, 17(4): 264-272.

[10] Xu G, Ma J, Liu F, et al. Automatic recognition of pavement surface crack based on BP neural network[C]//Proc of International conference on computer and electrical engineering. Piscataway, NJ: IEEE Press, 2008: 19-22.

[11] Jenkins M D, Carr T A, Iglesias M I, et al. A deep convolutional neural network for semantic pixel-wise segmentation of road and pavement surface cracks[C]//Proc of the 26th European signal processing conference (EUSIPCO). Piscataway, NJ: IEEE Press, 2018: 2120-2124.

[12] Gu Z, Chen H, Xu Z. Diffusioninst: Diffusion model for instance segmentation[C]// International Conference on Acoustics, Speech and Signal Processing (ICASSP). Piscataway, NJ: IEEE Press, 2024: 2730-2734.

[13] Nguyen Q, Vu T, Tran A, et al. Dataset diffusion: Diffusion-based synthetic data generation for pixel-level semantic segmentation [J]. Advances in Neural Information Systems, 2024, 36.

[14] Fang H, Han B, Zhang S, et al. Data augmentation for object detection via controllable diffusion models[C]//Proc of the IEEE/CVF Winter Conference on Applications of Computer Vision. 2024: 1257-1266.

[15] Trabucco B, Doherty K, Gurinas M, et al. Effective data augmentation with diffusion models[J]. arxiv preprint arxiv:2302.07944, 2023.

[16] Yu X, Li G, Lou W, et al. Diffusion-based data augmentation for nuclei image segmentation[C]// Proc of International Conference on Medical Image Computing and Computer-Assisted Intervention. Cham: Springer Nature Switzerland, 2023: 592-602.

[17] Song J, Meng C, Ermon S. Denoising diffusion implicit models[J]. arxiv preprint arxiv:2010. 02502, 2020.

[18] Woo S, Park J, Lee J Y, et al. Cbam: Convolutional block attention module[C]//Proc of the European conference on computer vision (ECCV). 2018: 3-19.

[19] Ronneberger O, Fischer P, Brox T. U-net: Convolutional networks for biomedical image segmentation[C]//Proc of the 18th Medical image computing and computer-assisted intervention–MICCAI international conference. Munich, Germany: Springer International, 2015: 234-241.

[20] Yang F, Zhang L, Yu S, et al. Feature pyramid and hierarchical boosting network for pavement crack detection [J]. IEEE Transactions on Intelligent Transportation Systems, 2019, 21(4): 1525-1535.