Awesome-Crack-Detection

📚About

This repository is largely based on the following paper:

Deep Learning for Crack Detection: A Review of Learning Paradigms, Generalizability, and Datasets
Xinan Zhang, Haolin Wang, Yung-An Hsieh, Zhongyu Yang, Anthony Yezzi, Yi-Chang Tsai

If you find this repository helpful, please consider giving a star and citing:

@article{zhang2025deep,
  title={Deep Learning for Crack Detection: A Review of Learning Paradigms, Generalizability, and Datasets},
  author={Zhang, Xinan and Wang, Haolin and Hsieh, Yung-An and Yang, Zhongyu and Yezzi, Anthony and Tsai, Yi-Chang},
  journal={arXiv preprint arXiv:2508.10256},
  year={2025}
}

🔥 News

• [2026-07-20] 🚀 We released a dataset for DL-based crack detection 3DCrack
• [2025-12-30] 🧑‍💻 We released a codebase for DL-based crack detection GT-CrackSeg
• [2025-08-14] 📢 Check out our comprehensive review paper Deep Learning for Crack Detection: A Review of Learning Paradigms, Generalizability, and Datasets
• [2025-08-12] We curated this list and published the first version.

Overview

• Review Papers
• Datasets
• Supervised Learning
• Semi-Supervised Learning
• Weakly Supervised Learning
• Domain Adaptation
• Few shot Learning
• Unsupervised Learning
• Foundation Models
• Citation

Review Papers

General CV

1. Review the state-of-the-art technologies of semantic segmentation based on deep learning, Neurocomputing, 2022. [Paper Link]

2. A Survey on Semi-Supervised Semantic Segmentation, Arxiv, 2023. [Paper Link]

3. A Survey of Weakly-supervised Semantic Segmentation, IEEE International Conference on Big Data Security on Cloud (BigDataSecurity), High Performance and Smart Computing (HPSC) and Intelligent Data and Security (IDS), 2023. [Paper Link]

4. Deep Unsupervised Domain Adaptation: A Review of Recent Advances and Perspectives, APSIPA Transactions on Signal and Information Processing, 2022. [Paper Link]

5. Few Shot Semantic Segmentation: a review of methodologies, benchmarks, and open challenges, Arxiv, 2024. [Paper Link]

Domain Specific

1. Crack detection using image processing: A critical review and analysis, Alexandria Engineering Journal, 2018. [Paper Link]

   • Literature presents different techniques to automatically identify the crack and its depth using image processing techniques. In this research, a detailed survey is conducted to identify the research challenges and the achievements till in this field. Accordingly, 50 research papers are taken related to crack detection, and those research papers are reviewed. Based on the review, analysis is provided based on the image processing techniques, objectives, accuracy level, error level, and the image data sets. Finally, we present the various research issues which can be useful for the researchers to accomplish further research on the crack detection.
   • 

2. Pavement crack image acquisition methods and crack extraction algorithms: A review, Journal of Traffic and Transportation Engineering, 2019. [Paper Link]

   • in order to know the developing history and the advanced research, we have collected a number of literature in this research topic for summarizing the research artwork status, and giving a review of the pavement crack image acquisition methods and 2D crack extraction algorithms. Also, for image acquisition methods and pavement crack image segmentation, more detailed comparison and discussions are made.
   • 

3. Review of Pavement Defect Detection Methods, IEEE Access, 2020. [Paper Link]

   • This paper reviews the three major types of methods used in road cracks detection: image processing, machine learning and 3D imaging based methods.
   • In this work, we review and compare the deep learning neural networks proposed in crack detection in three ways, classification based, object detection based and segmentation based.
   • 

4. Machine learning for crack detection: Review and model performance comparison, Journal of Computing in Civil Engineering, 2020. [Paper Link]

   • In this paper, the authors organize and provide up-to-date information on on ML-based crack detection algorithms for researchers to more efficiently seek potential focus and direction. The authors first reviewed 68 ML-based crack detection methods to identify the current trend of development, pixel-level crack segmentation. The authors then conducted a performance evaluation on 8 ML-based crack segmentation models using consistent evaluation metrics and three-dimensional (3D) pavement images with diverse conditions to identify remaining challenges and potential directions for future development.
   • 

5. Image-Based Crack Detection Methods: A Review, Infrastructures, 2021. [Paper Link]

   • Using image processing techniques, the captured or scanned images of the infrastructure parts can be analyzed to identify any possible defects. Apart from image processing, machine learning methods are being increasingly applied to ensure better performance outcomes and robustness in crack detection. This paper provides a review of image-based crack detection techniques which implement image processing and/or machine learning. A total of 30 research articles have been collected for the review which is published in top tier journals and conferences in the past decade. A comprehensive analysis and comparison of these methods are performed to highlight the most promising automated approaches for crack detection
   • 

6. Review on computer vision-based crack detection and quantification methodologies for civil structures, Construction and Building Materials, 2022. [Paper Link]

   • this review provides a comprehensive overview of state-of-the-art image-based crack analysis under various conditions in both qualitative and quantitative aspects, particularly focusing on image processing and deep learning-based methodologies from image-level detection to pixel-level segmentation and quantification.
   • The key challenges and research gaps are also discussed as follows, which indicate the importance of future research: (1) developing data model methodologies to resolve the difficulties due to the image data deficiency; (2) building a learning-based model capable of processing data with complex backgrounds; (3) enhancing the scene generalisation on different detection tasks; (4) establishing a lightweight mechanism for real-time crack analysis; (5) constructing learning-based systems that comprehend the local and global contexts during crack evaluation; (6) developing a semi-supervised mechanism for more information capturing and (7) establishing attention-based models for enhanced segmentation performance.
   • 

7. Deep learning-based crack segmentation for civil infrastructure: data types, architectures, and benchmarked performance, Automation in Construction, 2023. [Paper Link]

   • This paper reviews recent developments in deep learning-based crack segmentation methods and investigates their performance under the impact from different image types.
   • an image dataset, namely the Fused Image dataset for convolutional neural Network based crack Detection (FIND), was released to the public for deep learning analysis.
   • 

8. Computer vision framework for crack detection of civil infrastructure—A review, Engineering Applications of Artificial Intelligence, 2023. [Paper Link]

   • This paper provides a comprehensive review of the research progress and prospects in computer vision frameworks for crack detection of civil infrastructures from multiple materials, including asphalt, concrete, and metal-like materials. The review encompasses major components of typical frameworks, i.e., data acquisition techniques, publicly available datasets, detection algorithms, and evaluation metrics.
   • In particular, we provide a taxonomy of detection algorithms with a detailed discussion of the advantages, limitations, and application scenarios of the methods in each category, as well as the relationships between methods of different categories. We also discuss unsolved issues and key challenges in crack detection that could drive future research directions.
   • 

9. A Review of Computer Vision-Based Crack Detection Methods in Civil Infrastructure: Progress and Challenges, Remote Sens, 2024. [Paper Link]

   • Based on the main research methods of the 120 documents, we classify them into three crack detection methods: fusion of traditional methods and deep learning, multimodal data fusion, and semantic image understanding. We examine the application characteristics of each method in crack detection and discuss its advantages, challenges, and future development trends.
   • 

10. A critical review and comparative study on image segmentation-based techniques for pavement crack detection, Construction and Building Materials, 2022. [Paper Link]

    • This literature review establishes the history of development and interpretation of existing studies before conducting new research; and focuses heavily on three major types of approaches in the field of image segmentation, namely thresholding-based, edge-based, and data driven-based methods. With comparison and analysis of various image segmentation algorithms, this research provides valuable information for researchers working on enhanced segmentation strategies that potentially yield a fully automated distress detection process for pavement images with varying conditions.
    • 

11. A review of deep learning methods for pixel-level crack detection, Journal of Traffic and Transportation Engineering (English Edition), 2022. [Paper Link]

    • we present a comprehensive thematic survey of DL-based CIS techniques. Our review offers several contributions to the CIS area. First, more than 40 papers of journal or top conference most published in the last three years are identified and collected based on the systematic literature review method. Second, according to the backbone network architecture of the models proposed in them, they are grouped into 10 topics: FCN, U-Net, encoder-decoder model, multi-scale, attention mechanism, transformer, two-stage detection, multi-modal fusion, unsupervised learning and weakly supervised learning, to be reviewed. Meanwhile, our survey focuses on discussing strengths and limitations of the models in each topic so as to reveal the latest research progress in the CIS field. Third, publicly accessible data sets, evaluation metrics, and loss functions that can be used for pixel-level crack detection are systematically introduced and summarized to facilitate researchers to select suitable components according to their own research tasks. Finally, we discuss six common problems and existing solutions to them in the field of DL-based CIS, and then suggest eight possible future research directions in this field.
    • 

12. A Comprehensive Review of Deep Learning-Based Crack Detection Approaches, Appl. Sci., 2022. [Paper Link]

    • a comprehensive literature review of deep learning-based crack detection studies and the contributions they have made to the field is presented. The studies are categorised according to the computer vision aspect and at deeper levels to facilitate exploring the studies that utilised similar approaches to address the crack detection problem. Moreover, the authors perform a comparison between the studies which use the same publicly available data sets, in order to find the most promising crack detection approaches. Critical future directions for research are proposed, based on these reviewed studies as well as on trends and developments in areas similar to the crack detection area.
    • 

13. Structural crack detection using deep convolutional neural networks, Automation in Construction, 2022. [Paper Link]

    • This article presents a review of CNN implementation on civil structure crack detection. The review highlights the significant research that has been performed to detect structure cracks through classification and segmentation of crack images with CNN in the perspective of image pre-processing techniques, processing hardware, software tools, datasets, network architectures, learning procedures, loss functions, and network performance.
    • 

14. Deep Learning-Based Crack Detection: A Survey, *International Journal of Pavement Research and Technology *, 2022. [Paper Link]

    • This study has identified the bigger picture of DL methods for crack identification in asphalt pavement. The authors evaluated several DL-based crack identification algorithms from the literature, such as crack classification, crack object detection, pixel-level crack segmentation, generative adversarial networks (GANs) for crack segmentation, and crack identification using unsupervised learning. Moreover, 26 DL-based crack detection models (25 supervised learning models and one unsupervised learning model) were analysed on the same dataset to test the performance of each model using consistent assessment metrics. The testing results suggest that ResNet and DenseNet are the best options for crack classification, while Faster R-CNN should be used for crack object detection and pix2pix is suggested for crack segmentation. It is also recommended that semi-supervised and unsupervised learning be further studied to efficiently detect cracks in an asphalt pavement.
    • 

15. Visual Concrete Bridge Defect Classification and Detection Using Deep Learning: A Systematic Review, IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS, 2024. [Paper Link]

    • Our aim in this survey is to study the recent progress of vision-based concrete bridge defect classification and detection in the deep learning era. Our review encompasses major aspects underlying typical frameworks, which include concrete defect taxonomy, public datasets and evaluation metrics.
    • 

16. Few-shot learning for structural health diagnosis of civil infrastructure, Advanced Engineering Informatics, 2024. [Paper Link]

    • This article systematically summarizes recent advances in FSL algorithms and the corresponding applications in SHD for civil infrastructure. A unified mathematical framework of FSL is formulated, and an FSL taxonomy is summarized according to intrinsic learning mechanisms and implementation principles, including metric learning-based, optimization-based, transfer learning-based, and generative model-based methods. Various applications of SHD for civil infrastructure under real-world scenarios are reviewed, including remote sensing monitoring, structural damage recognition, post-disaster safety evaluation, and construction risk assessment.
    • 

Datasets

1. CrackTree200: Automatic crack detection from pavement images, Pattern Recognition Letters, 2012. [Paper Link]

   • We collect a set of 206 pavement images with various kinds of cracks. All these images have a size of 800 × 600, and many of them suffer from the problems of shadows, occlusions, low contrast, noise, etc.
   • 

2. CrackIT — An image processing toolbox for crack detection and characterization, ICIP, 2014. [Paper Link]

   • A sample database of 84 pavement surface images taken during a traditional road survey is provided with the toolbox, since no pavement image databases are publicly available for crack detection and characterization evaluation purposes.
   • 

3. CFD: Automatic Road Crack Detection Using Random Structured Forests, IEEE Transactions on Intelligent Transportation Systems, 2016. [Paper Link]

   • We propose an annotated road crack dataset called CFD. This dataset is composed of 118 images, which can generally reflect urban road surface condition in Beijing, China. Each image has hand labeled ground truth contours. All the images are taken by an iPhone5 with focus of 4mm, aperture of f/2.4 and exposure time of 1/134s. The width of the images ranges from 1 to 3 mm.
   • 

4. CRACK500: Road crack detection using deep convolutional neural network, ICIP, 2016. [Paper Link]

   • Data set with more than 500 pavement pictures of size 3264×2448 are collected at the Temple University campus by using a smart phone as the data sensor. Each image is annotated by multiple annotators.
   • 

5. AEL: Automatic Crack Detection on Two-Dimensional Pavement Images: An Algorithm Based on Minimal Path Selection, IEEE Transactions on Intelligent Transportation Systems, 2016. [Paper Link]

   • This paper proposes a new algorithm for automatic crack detection from 2D pavement images. It strongly relies on the localization of minimal paths within each image, a path being a series of neighboring pixels and its score being the sum of their intensities. The originality of the approach stems from the proposed way to select a set of minimal paths and the two postprocessing steps introduced to improve the quality of the detection.
   • An intensive validation is performed on both synthetic and real images (from five different acquisition systems), with comparisons to five existing methods.
   • 

6. CSSC: Deep Concrete Inspection Using Unmanned Aerial Vehicle Towards CSSC Database, IROS, 2017. [Paper Link]

   • This paper presents an automated approach using Unmanned Aerial Vehicle(UAV) and towards a Concrete Structure Spalling and Crack database (CSSC), which is by far the first released database for deep learning inspection
   • 

7. FCN: Automatic Pixel-Level Crack Detection and Measurement Using Fully Convolutional Network, Computer-Aided Civil and Infrastructure Engineering, 2018. [Paper Link]

   • To train the FCN model, the authors collected more than 800 images. The width of cracks varies from one pixel to 100 pixels, the shape of which is hard to recognize at image level. To ensure the variability, historical cracks from internet and new cracks from existing buildings in Harbin, China, are involved. Not only pavement cracks, but also cracks on concrete walls are contained and saved in JPG format. Cracks in these images are taken at different distances depending on their sizes, which leads to different levels of resolutions, ranging from 72 dpi to 300 dpi.
   • 

8. CRKWH100, CrackTree260, CrackLS315, Stone331: DeepCrack: Learning Hierarchical Convolutional Features for Crack Detection, IEEE Transactions on Image Processing, 2018. [Paper Link]

   • CRKWH100
   • CrackTree260
   • CrackLS315: Images in this dataset are captured under laser illumination, which makes them more different with the training images than that in CRKWH100
   • Stone331
   • 

9. DeepCrack: DeepCrack: A deep hierarchical feature learning architecture for crack segmentation, Neurocomputing, 2019. [Paper Link]

   • A benchmark dataset consisting of 537 images with manual annotation maps are built
   • 

10. GAPS384: Feature Pyramid and Hierarchical Boosting Network for Pavement Crack Detection, IEEE Transactions on Intelligent Transportation Systems, 2019. [Paper Link]

    • we manually select 384 images from the GAPs dataset, which only includes crack class of distress, and conduct pixel-wise annotation. This pixel-wise annotated crack dataset is named as GAPs384
    • 

11. KolektorSDD: Segmentation-based deep-learning approach for surface-defect detection, Journal of Intelligent Manufacturing, 2019. [Paper Link]

    • An extensive evaluation of the proposed method is performed on a novel, real-world dataset termed Kolektor Surface-Defect Dataset (KolektorSDD). The dataset represents a real-world problem of surface-defect detection for an industrial semi-finished product where the number of defective items available for the training is limited.
    • 

12. Khanh11k: crack-segmentation, github, 2019. [Paper Link]

    • It contains around 11.200 images that are merged from 12 available crack segmentation datasets.
    • 

13. UAV75: Crack Segmentation on UAS-based Imagery using Transfer Learning, International Conference on Image and Vision Computing New Zealand (IVCNZ), 2019. [Paper Link]

    • The dataset called UAV75, cf. Table I, contains 75 manually annotated images of size 512×512 pix with pixel-wise labels. The images were manually split into training, validation, and test dataset, each subset covering the perceptible variations in crack and planking patterns.

14. DIC: Comparison of crack segmentation using digital image correlation measurements and deep learning, Construction and Building Materials, 2020. [Paper Link]

    • For the training and validation data (dataset A), 17 full-size laboratory images were selected from the specimens RSC2-3, RSD1, RSD2, RSM2, SC1-7, RS1-3. The images taken during the experimental campaigns illustrate the state of the plastered wall surfaces under different loading levels and support conditions
    • For the test data (dataset B), three full-size images were selected from specimens RS4 and RS6 at different loading levels (see Fig. 5). To obtain a fair model, no images taken from specimens RS4 and RS6 were included in the training/validation data (dataset A)
    • 

15. Masonry: Automatic crack classification and segmentation on masonry surfaces using convolutional neural networks and transfer learning, Automation in Construction, 2021. [Paper Link]

    • A dataset with photos from masonry structures is produced containing complex backgrounds and various crack types and sizes.
    • 

16. Ceramic: Ceramic Cracks Segmentation with Deep Learning, applied sciences, 2021. [Paper Link]

    • This work focuses on automated optical inspection to find faults in ceramic tiles performing the segmentation of cracks in ceramic images using deep learning to segment these defects. We propose an architecture for segmenting cracks in facades with Deep Learning that includes an image pre-processing step. We also propose the Ceramic Crack Database, a set of images to segment defects in ceramic tiles.
    • 

17. CrSpEE: DETECTING CRACKS AND SPALLING AUTOMATICALLY IN EXTREME EVENTS BY END-TO-END DEEP LEARNING FRAMEWORKS, ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2021. [Paper Link]

    • A total of 2,229 images were annotated, for cracking and spalling
    • 

18. BCL: Structural Crack Detection from Benchmark Data Sets Using Pruned Fully Convolutional Networks, Journal of Structural Engineering, 2021. [Paper Link]

    • In this study, a benchmark data set called a bridge crack library (BCL) containing 11,000 pixel-wise labeled images with 256×256 resolution was established, which has 5,769 nonsteel crack images, 2,036 steel crack images, 3,195 noise images, and their labels. It is aimed at crack detection on multiple structural materials including masonry, concrete, and steel. The raw images were collected by multiple cameras from more than 50 in-service bridges during a period of 2 years.
    • 
    • 

19. Conglomerate: Development of Extendable Open-Source Structural Inspection Datasets, Journal of Computing in Civil Engineering, 2022. [Paper Link]

    • we set out to acquire bridge inspection data by selectively extracting candidate images from hundreds of thousands of bridge inspection reports from the Virginia Department of Transportation. Using this rich source of diverse data, we refined our collected data to develop four high-quality, easily extendable, publicly accessible datasets, tested with state-of-the-art models to support typical bridge inspection tasks. The four datasets: labeled cracks in the wild, 3,817 image sets of semantically segmented concrete cracks taken from diverse scenery; 3,817 image sets of semantically segmented structural inspection materials (concrete, steel, metal decking)
    • The conglomerate dataset comprises 10,995 images that have been merged from CFD (Shi et al. 2016), Crack500 (Yang et al. 2020), CrackTree200 (Zou et al. 2012), DeepCrack (Liu et al. 2019), Eugen Miller (Yang et al. 2020), GAPs (Eisenbach et al. 2017), Rissbilder (Yang et al. 2020), noncrack (Dorafshan et al. 2018), and Volker (Yang et al. 2020). The dataset details are summarized in Table
    • 
    • 

20. LCW: Development of Extendable Open-Source Structural Inspection Datasets, Journal of Computing in Civil Engineering, 2022. [Paper Link]

    • The LCW dataset comprised images that were taken from VDOT Bridge Inspection Reports, showing a more global scene, rather than focused completely on concrete
    • 

21. Syncrack: Improving Pavement and Concrete Crack Detection through Synthetic Data Generation, VISAPP, 2022. [Paper Link]

    • manual annotations tend to be inaccurate, particularly at pixel-accurate level. The learning bias introduced by this inaccuracy hinders pixel-accurate crack detection. In this paper we propose a novel tool aimed for synthetic image generation with accurate crack labels – Syncrack. This parametrizable tool also provides a method to introduce controlled noise to annotations, emulating human inaccuracy
    • 

22. TopoDS: TOPO-Loss for continuity-preserving crack detection using deep learning, Construction and Building Materials, 2022. [Paper Link]

    • New training dataset of real-world post-earthquake building images containing labeled cracks
    • 

23. CrackSeg9k: A Collection and Benchmark for Crack Segmentation Datasets and Frameworks, ECCV W, 2022. [Paper Link]

    • This paper addresses these problems by combining previously available datasets and unifying the annotations by tackling the inherent problems within each dataset, such as noise and distortions.
    • We also present a pipeline that combines Image Processing and Deep Learning models. Finally, we benchmark the results of proposed models on these metrics on our new dataset and compare them with state-of-the-art models in the literature.
    • 

24. S2DS: Image-Based Detection of Structural Defects Using Hierarchical Multi-scale Attention, DAGM GCPR, 2022. [Paper Link]

    • The dataset of 743 images covering crack, spalling, corrosion, efflorescence, vegetation, and control point is unprecedented in terms of quantity and realism
    • 

25. FIND: Deep Learning-Based Crack Detection and Segmentation Using High-Resolution 3D Laser Imaging, Automation in Construction, 2023. [Paper Link]

    • The FIND dataset consists of 2,500 crack image patches collected from bridge decks and roadways using a high-resolution laser scanning system. Each image includes surface elevation information and comes in four types: raw intensity, raw range, filtered range, and fused representations.
    • 

26. CrackMap: A Few‑Shot Attention Recurrent Residual U‑Net for Crack Segmentation, arXiv, 2023. [Paper Link]

    • CrackMap contains 120 road crack images collected using a vehicle-mounted RGB camera, representing typical road scenes. The dataset is designed to support few-shot learning experiments in crack detection.
    • 

27. TUT Crack Dataset: Staircase Cascaded Fusion for Structural Crack Segmentation, arXiv, 2024. [Paper Link]

    • The TUT dataset includes 1,408 crack images, with 1,270 collected via mobile phones and 138 from online sources. It features diverse, real-world backgrounds and complex crack shapes, offering a challenging testbed for crack segmentation.
    • 

28. SegCODEBRIM: Hybrid Neural System to Learn Real-World Crack Segmentation, WACV, 2024. [Paper Link]

    • SegCODEBRIM includes 420 crack segmentation masks added to the CODEBRIM dataset, originally a collection of concrete bridge defect images. It enables realistic training and evaluation for deep learning-based crack detection on concrete surfaces.
    • 

29. Crack900: Crack detection of masonry structure based on thermal and visible image fusion and semantic segmentation, Automation in Construction, 2023. [Paper Link]

    • 914 sets of visible and infrared (IR) images captured from masonry walls using solar heating as the sole heat source.
    • 

30. MCrack1300: Sam-based instance segmentation models for the automation of structural damage detection, Advanced Engineering Informatics, 2024. [Paper Link]

    • 1,300 masonry crack images sourced from Crack900, online images, and mobile phone photos, covering diverse brick types and crack patterns.
    • 

31. OMNICRACK30K: A Benchmark for Crack Segmentation and the Reasonable Effectiveness of Transfer Learning, CVPR W, 2024. [Paper Link]

    • The OMNICRACK30K dataset forms the first large-scale, systematic, and thorough approach to provide a sustainable basis for tracking methodical progress in the field of crack segmentation. It contains 30k samples from over 20 datasets summing up to 9 billion pixels in total. Featuring materials as diverse as asphalt, ceramic, con- crete, masonry, and steel, it paves the road towards univer- sal crack segmentation, a currently under-explored topic.

    • 

    • Tab. 4 shows the results of the benchmarking approaches on the test subsets of OMNICRACK30K. Note that only the general models are trained on OMNICRACK30K, while the crack-specific approaches used their specific datasets.

    • 

Learning Paradigms

Supervised Learning

1. Automated Pixel-Level Pavement Crack Detection on 3D Asphalt Surfaces Using a Deep-Learning Network, Computer-Aided Civil and Infrastructure Engineering, 2017. [Paper Link]

   • The CrackNet, an efficient architecture based on the Convolutional Neural Network (CNN), is proposed in this article for automated pavement crack detection on 3D asphalt surfaces with explicit objective of pixel-perfect accuracy. Unlike the commonly used CNN, CrackNet does not have any pooling layers which downsize the outputs of previous layers.
   • The input data of the CrackNet are feature maps generated by the feature extractor using the proposed line filters with various orientations, widths, and lengths. The output of CrackNet is the set of predicted class scores for all pixels.
   • 

2. Deep Learning–Based Fully Automated Pavement Crack Detection on 3D Asphalt Surfaces with an Improved CrackNet, Journal of Computing in Civil Engineering, 2018. [Paper Link]

   • This paper proposes an improved architecture of CrackNet called CrackNet II for enhanced learning capability and faster performance. The proposed CrackNet II represents two major modifications on the original CrackNet.
   • First, the feature generator, which provides handcrafted features through fixed and nonlearnable procedures, is no longer used in CrackNet II. Consequently, all layers in CrackNet II have learnable parameters that are tuned during the learning process. Second, CrackNet II has a deeper architecture with more hidden layers but fewer parameters.
   • Compared with the original CrackNet, CrackNet II is capable of detecting more fine or hairline cracks, while eliminating more local noises and maintaining much faster processing speed.
   • 

3. DeepCrack: Learning Hierarchical Convolutional Features for Crack Detection, IEEE Transactions on Image Processing, 2018. [Paper Link]

   • In this paper, we propose DeepCrack-an end-to-end trainable deep convolutional neural network for automatic crack detection by learning high-level features for crack representation. In this method, multi-scale deep convolutional features learned at hierarchical convolutional stages are fused together to capture the line structures
   • We build DeepCrack net on the encoder–decoder architecture of SegNet and pairwisely fuse the convolutional features generated in the encoder network and in the decoder network at the same scale
   • 

4. Automated Pixel-Level Pavement Crack Detection on 3D Asphalt Surfaces with a Recurrent Neural Network, Computer-Aided Civil and Infrastructure Engineering, 2018. [Paper Link]

   • A recurrent neural network (RNN) called CrackNet-R is proposed in the article for fully automated pixel-level crack detection on three-dimensional (3D) asphalt pavement surfaces. In the article, a new recurrent unit, gated recurrent multilayer perceptron (GRMLP), is proposed to recursively update the internal memory of CrackNet-R
   • CrackNet-R implements a two-phase sequence processing: sequence generation and sequence modeling. Sequence generation is specifically developed in the study to find the best local paths that are most likely to form crack patterns. Sequence modeling predicts timely probabilities of the input sequence being a crack pattern
   • 

5. DeepCrack: A deep hierarchical feature learning architecture for crack segmentation, Neurocomputing, 2019. [Paper Link]

   • In this paper, we propose a deep hierarchical convolutional neural network (CNN), called as DeepCrack, to predict pixel-wise crack segmentation in an end-to-end method.
   • DeepCrack consists of the extended Fully Convolutional Networks (FCN) and the Deeply-Supervised Nets (DSN). During the training, the elaborately designed model learns and aggregates multi-scale and multi-level features from the low convolutional layers to the high-level convolutional layers, which is different from the standard approaches of only using the last convolutional layer
   • 

6. Pavement crack detection and recognition using the architecture of segNet, Journal of Industrial Information Integration, 2020. [Paper Link]

   • We propose an encoder-decoder structural model with a fully convolutional neural network, namely, PCSN, by referring to SegNet.
   • 

7. DAU-Net: Dense Attention U-Net for Pavement Crack Segmentation, International Conference on Intelligent Transportation, 2021.
   [Paper Link]

   • we proposed the Dense Attention U-Net (DAU-Net) to achieve pixel-wise segmentation of cracks on 3D pavement images.
   • The encoder of the DAU-Net consists of multi-stage dense blocks to improve its capability of extracting informative contextual features.
   • To achieve precise localization of cracks in the decoder, a novel channel attention block (CAB) is proposed, which reduces noisy responses and highlight salient encoder features using the channel attention mechanism.
   • 

8. A Crack Detection Algorithm for Concrete Pavement Based on Attention Mechanism and Multi-Features Fusion, IEEE Transactions on Intelligent Transportation Systems, 2021. [Paper Link]

   • Inspired by the latest developments of deep learning in computer vision, we propose a novel crack detection algorithm of concrete pavement based on attention mechanism and multi-features fusion, and make it possible to deal with various cracks in different pavement backgrounds.
   • The proposed network is constructed using the encoder-decoder structure. The architecture of the encoder part is consisted of Res2Net modules with attention mechanism to achieve fast focus of cracks. Cascade and parallel mode dilated convolutions are set as the center part to enlarge the receptive field of feature points without reducing the resolution of the feature maps.
   • 

9. SDDNet: Real-Time Crack Segmentation, IEEE Transactions on Industrial Electronics, 2021. [Paper Link]

   • The model consists of standard convolutions, densely connected separable convolution modules, a modified atrous spatial pyramid pooling module, and a decoder module.
   • In addition, the model processes in real-time (36 FPS) images at 1025 × 512 pixels, which is 46 times faster than in a recent work
   • 

10. Efficient attention-based deep encoder and decoder for automatic crack segmentation, Structural Health Monitoring, 2021. [Paper Link]

    • In this paper, a novel semantic transformer representation network (STRNet) is developed for crack segmentation at the pixel level in complex scenes in a real-time manner. STRNet is composed of a squeeze and excitation attention-based encoder, a multi head attention-based decoder, coarse upsampling, a focal-Tversky loss function, and a learnable swish activation function to design the network concisely by keeping its fast-processing speed.
    • 

11. CrackFormer: Transformer Network for Fine-Grained Crack Detection, ICCV, 2021. [Paper Link]

    • We propose a Crack Transformer network (CrackFormer) for fine-grained crack detection. The CrackFormer is composed of novel atten- tion modules in a SegNet-like encoder-decoder architecture.
    • Specifically, it consists of novel self-attention modules with 1x1 convolutional kernels; It also introduces new scaling-attention modules to combine outputs from the corresponding encoder and decoder blocks to suppress non- semantic features and sharpen semantic ones.
    • 

12. Pavement crack detection based on transformer network, Automation in Construction, 2023.
    [Paper Link]

    • This study aims to improve pavement crack detection under noisy conditions.
    • A novel model named Crack Transformer (CT), which unifies Swin Transformer as the encoder and the decoder with all multi-layer perception (MLP) layers, is proposed for the automatic detection of long and complicated pavement cracks.
    • This study shows the feasibility of using a Transformer-based network for automatic robust pavement crack detection under noisy conditions.
    • 

13. A Convolutional-Transformer Network for Crack Segmentation with Boundary Awareness, ICIP, 2023. [Paper Link]

    • we propose a novel convolutional-transformer network based on encoder-decoder architecture to solve this challenge. Particularly, we designed a Dilated Residual Block (DRB) and a Boundary Awareness Module (BAM). The DRB pays attention to the local detail of cracks and adjusts the feature dimension for other blocks as needed. And the BAM learns the boundary features from the dilated crack label.
    • Furthermore, the DRB is combined with a lightweight transformer that captures global information to serve as an effective encoder
    • 

14. Crack detection and quantification for concrete structures using UAV and transformer, Automation in Construction, 2023.
    [Paper Link]

    • Proposes a UAV-based approach for accurate detection and quantification of concrete cracks without reference markers
    • Introduces an Independent Boundary Refinement Transformer (IBR-Former) for crack segmentation from UAV-captured images
    • Achieves quantification of cracks with widths less than 0.2 mm, meeting civil engineering code requirements
    • 

15. Selective Feature Fusion and Irregular-Aware Network for Pavement Crack Detection, IEEE Transactions on Intelligent Transportation Systems, 2023. [Paper Link]

    • this paper introduces an innovative neural network architecture termed the ‘Selective Feature Fusion and Irregular-Aware Network (SFIAN)’ designed specifically for crack detection on pavements. The proposed network selectively integrates features from multiple levels, enhancing and controlling the flow of valuable information at each stage while effectively modeling irregular crack objects
    • 

16. A crack-segmentation algorithm fusing transformers and convolutional neural networks for complex detection scenarios, Automation in Construction, 2023. [Paper Link]

    • a dual-encoder network fusing transformers and convolutional neural networks (DTrC-Net) is proposed in this study. The structure of the DTrC-Net was designed to capture both the local features and global contextual information of crack images. To enhance feature fusion between the adjacent and codec layers, a feature fusion module and a residual path module were also added to the network
    • Moreover, a fast processing speed of 78 frames per second was achieved using the DTrC-Net with an image size of 256 × 256 pixels
    • 

17. Dual-path network combining CNN and transformer for pavement crack segmentation, Automation in Construction, 2024. [Paper Link]

    • a dual-path network for pavement crack segmentation is introduced, leveraging a synergistic combination of Convolutional Neural Network (CNN) and transformer.
    • First, the proposed approach involves a lightweight CNN encoder for local feature extraction and a novel transformer encoder integrating a fully convolutional high-low frequency attention (FCHiLo) mechanism and an efficient feedforward network for global feature extraction. Second, a complementary fusion module (CFM) is introduced to aggregate intermediate features extracted from both encoders.
    • 

18. Staircase Cascaded Fusion of Lightweight Local Pattern Recognition and Long-Range Dependencies for Structural Crack Segmentation, arXiv, 2024. [Paper Link]

    • we propose the Staircase Cascaded Fusion Crack Segmentation Network (CrackSCF), which generates high-quality crack segmentation maps while reducing computational overhead.
    • We design a lightweight convolutional block that substitutes all convolution operations, reducing the model's computational demands while maintaining an effective capture of local details. Additionally, we introduce a lightweight long-range dependency extractor to better capture the long-range dependencies. Furthermore, we develop a staircase cascaded fusion module, which seamlessly integrates local patterns and long-range dependencies, resulting in high-quality segmentation maps.
    • The F1 and mIoU scores on the TUT dataset are 0.8382 and 0.8473, respectively, demonstrating state-of-the-art (SOTA) performance with low computational resources.
    • 

19. SCSegamba: Lightweight Structure-Aware Vision Mamba for Crack Segmentation in Structures, CVPR, 2025. [Paper Link]

    • we propose a lightweight Structure-Aware Vision Mamba Network (SCSegamba), capable of generating high-quality pixel-level segmentation maps by leveraging both the morphological information and texture cues of crack pixels with minimal computational cost.
    • Specifically, we developed a Structure-Aware Visual State Space module (SAVSS), which incorporates a lightweight Gated Bottleneck Convolution (GBC) and a Structure-Aware Scanning Strategy (SASS). The key insight of GBC lies in its effectiveness in modeling the morphological information of cracks, while the SASS enhances the perception of crack topology and texture by strengthening the continuity of semantic information between crack pixels.
    • Experiments on crack benchmark datasets demonstrate that our method outperforms other state-of-the-art (SOTA) methods, achieving the highest performance with only 2.8M parameters. On the multi-scenario dataset, our method reached 0.8390 in F1 score and 0.8479 in mIoU.
    • 

Semi-Supervised Learning

1. Semi-supervised semantic segmentation network for surface crack detection, Automation in Construction, 2021. [Paper Link]

   • we propose a semi-supervised semantic segmentation network for crack detection. The proposed method consists of student model and teacher model.
   • The two models have the same network structure and use the EfficientUNet to extract multi-scale crack feature information, reducing the loss of image information. The student model updates weights through the gradient descent of loss function, and the teacher model uses the exponential moving average weights of the student model. During training, the robustness of the model is improved by adding noise to the input data.
   • When using only 60% of the annotated data, our method achieves an F1 score of 0.6540 on the concrete crack dataset and 0.8321 on the Crack500 dataset.
   • 

2. Semi-supervised learning framework for crack segmentation based on contrastive learning and cross pseudo supervision, Measurement, 2023. [Paper Link]

   • A novel semi-supervised learning framework for crack segmentation, which is referred to as semi-supervised crack (SemiCrack), based on the combination of contrastive learning and cross pseudo supervision (CPS) is presented in this study.
   • The proposed segmentation network, called transformer and convolutional network (TC-Net), has a novel parallel encoder that fuses a transformer and a convolutional neural network. The inclusion of CPS can force the two models to maintain consistent outputs for various perturbed data based on the similarity loss. To capture the feature differences between positive and negative sample pairs extracted by the classifier and projector, pixel contrastive loss was also proposed.
   • SemiCrack requires only 20% labeled data to achieve comparable accuracy to other fully-supervised algorithms that require 100% labeled data.
   • 

3. End-to-end semi-supervised deep learning model for surface crack detection of infrastructures, Frontiers in Materials, 2022. [Paper Link]

   • we propose a novel semi-supervised learning model for crack detection. The proposed model employs a modified U-Net, which has half the parameters of the original U-Net network to detect surface cracks.
   • At each stage, the trained model predicts and segments the unlabeled data images. The new strategy for updating the training datasets allows the model to be trained with limited labeled image data.
   • Results show that the proposed semi-supervised learning method achieved quite approaching accuracies to the established fully supervised models using multiple accuracy indexes, however, the requirement for the labeled data reduces to 40%.
   • 

4. Cross teacher pseudo supervision: Enhancing semi-supervised crack segmentation with consistency learning, Advanced Engineering Informatics, 2024. [Paper Link]

   • To address this issue, a new semi-supervised algorithm is proposed to leverage both labeled and unlabeled data through a cross-teacher-pseudo-supervision framework and cross-augmentation strategy. The proposed method employs two pairs of teacher–student models to mutually supervise each other using pseudo-labels generated from their respective teacher models. To boost the performance of the proposed algorithm, input, feature, and network perturbances are applied during training.
   • Specifically, on the four datasets, the proposed method outperforms the supervised-only baseline by 0.92%, 1.29%, 4.14%, and 5.38% respectively, in mean intersection over union under the labeled ratio of 5%, and by 0.76%, 1.06%, 1.79% and 1.35% under the ratio of 10%
   • 

5. Pavement Crack Detection Using Fractal Dimension and Semi-Supervised Learning, Fractal Fract., 2014. [Paper Link]

   • This study proposes a pavement crack image detection method integrating fractal dimension analysis and semi-supervised learning. It identifies the self-similarity characteristics within the crack regions by analyzing pavement crack images and using fractal dimensions to preliminarily determine the candidate crack regions.
   • The Crack Similarity Learning Network (CrackSL-Net) is then employed to learn the semantic similarity of crack image regions. Semi-supervised learning facilitates automatic crack detection by combining a small amount of labeled data with a large volume of unlabeled image data.
   • The results indicate that, with a 50% annotation ratio, the proposed method achieves high-precision crack detection, with an intersection over union (IoU) exceeding 0.84, which is close to that of U-Net.
   • 

6. Semi-supervised crack detection using segment anything model and deep transfer learning, Automation in Construction, 2025. [Paper Link]

   • This paper proposes a semi-supervised instance segmentation method for road distress detection based on deep transfer learning. The interactive segmentation method utilizing SAM are used to enhance the production efficiency of segmentation datasets.
   • The DCNv3 and lightweight segmentation heads are strategically designed to offset potential speed losses. The deep transfer learning method fine-tunes the pre-trained models, enhancing their competency for new tasks.
   • The proposed model achieves comparable performance to supervised learning with fewer annotated data, accurately determining crack dimensions across varied scenarios
   • 

7. Semi-supervised semantic segmentation using cross-consistency training for pavement crack detection, *Road Materials and Pavement Design *, 2023. [Paper Link]

   • A semi-supervised semantic word segmentation method based on cross-consistency training for road crack detection is established in this paper. To take advantage of unlabelled crack samples, this study enforced consistency between the primary and secondary decoder predictions, using different disturbed versions of the encoder output as inputs to improve the encoder representation.
   • When only 60% of the annotated data were used, the applied method achieved a better performance than other mainstream semantic segmentation algorithms.
   • 

8. Automatic Road Crack Detection Using Convolutional Neural Network Based on Semi-Supervised Learning, * Engineering and Applied Sciences*, 2024. [Paper Link]

   • To address this challenge, this paper proposes a semi-supervised learning approach based on a DenseNet classification model to detect pavement cracks more efficiently.
   • The primary objective is to leverage a small set of labeled samples to improve the model's performance by incorporating a large number of unlabeled samples through semi-supervised learning. This method enhances the DenseNet model's ability to generalize by iteratively learning from new unlabeled datasets.
   • 

9. Semi-Supervised Semantic Segmentation Using Adversarial Learning for Pavement Crack Detection, IEEE Access, 2020. [[Paper Link(https://ieeexplore.ieee.org/abstract/document/9032091)]

   • To ease the workload of the inspector and lower the cost of acquiring the high-quality training dataset, a semi-supervised method for the pavement crack detection is proposed.
   • Firstly, unlabeled pavement images can be used for the model training in our proposed algorithm, our model can generate a supervisory signal for unlabeled pavement images, which makes up for the deficiency of image annotation. Secondly, an adversarial learning method and a full convolution discriminator are adopted, which can learn to distinguish the ground truth from segmentation predictions.
   • 

10. Cross teacher pseudo supervision: Enhancing semi-supervised crack segmentation with consistency learning, Advanced Engineering Informatics, 2024. [Paper Link]

    • a new semi-supervised algorithm is proposed to leverage both labeled and unlabeled data through a cross-teacher-pseudo-supervision framework and cross-augmentation strategy. The proposed method employs two pairs of teacher–student models to mutually supervise each other using pseudo-labels generated from their respective teacher models. To boost the performance of the proposed algorithm, input, feature, and network perturbances are applied during training
    • the proposed method outperforms the supervised-only baseline by 0.92%, 1.29%, 4.14%, and 5.38% respectively
    • 

11. Multiscale and adversarial learning-based semi-supervised semantic segmentation approach for crack detection in concrete structures, IEEE Access, 2020. [Paper Link]

    • To reduce these costs, in this study, multiscale and adversarial learning techniques were applied to realize crack detection. A total of 1,200 labeled data and 3,000 unlabeled data were used to implement and verify the proposed method. The multiscale segmentation neural network, discriminator neural network, and adversarial learning technique were used to realize accurate crack detection, enhance the learning performance, and ensure the efficiency of training data, respectively.
    • The resulting algorithm had a pixel accuracy, mean intersection over union, frequency weighted intersection over union, and F1 score of 98.176%, 88.936%, 96.525%, and 88.789%, respectively.
    • 

12. Efficient semi-supervised surface crack segmentation with small datasets based on consistency regularisation and pseudo-labelling, Automation in Construction, 2024. [Paper Link]

    • a semi-supervised framework is proposed, capable of learning from a substantial amount of unlabelled data and achieving high accuracy, even when the available labelled datasets are of limited size. The framework is designed by tailoring supervised training, semi-supervised consistency regularisation, and self-training with certainty-based pseudo-labelling, resulting in a simple yet effective approach.
    • the designed framework assisted the model in approaching and even exceeding saturation levels with as little as 20% and 25% of the Concrete and Asphalt datasets.
    • 

Weakly Supervised

1. Patch-based weakly supervised semantic segmentation network for crack detection, Construction and Building Materials, 2020. [Paper Link]

   • this paper proposes a patch-based weakly supervised semantic segmentation network for crack detection. The proposed method uses image-level annotation as the supervision condition and fully considers the local similarity of the crack topology in the image. The use of patches cropped from the image as the input in this method can reduce the image complexity significantly without losing the spatial location information of the crack.
   • A discriminative localization technique is used to extract rough location information of the crack from a trained classification network, which is then refined by a conditional random field (DenseCRF) to obtain a synthetic label. These synthetic labels can replace the manually annotated pixel-level labels for the training of the segmentation network. Thereafter, a neighborhood fusion strategy is used to merge the patches into the final output.
   • reducing the annotation workload by approximately 80%
   • 

2. A weakly-supervised transformer-based hybrid network with multi-attention for pavement crack detection, Construction and Building Materials, 2024. [Paper Link]

   • we propose a novel Weakly-Supervised hybrid network with multi-attention, termed CGTr-Net, for pavement crack detection.
   • Aiming at alleviating the loss of information, behaving well in extracting both local and global features, the architecture of the backbone CG-Trans was designed. It is a combination of Convolutional Neural Network (CNN), which is expert in extracting local features but experiencing difficulties to capture global representations, and Gated axial Transformer, whose gated position-sensitive axial attention mechanism can efficiently extract long-distance feature dependencies but deteriorate in capturing local feature details.
   • To enhance feature fusion between the Transformer Layer and the Convolution Layer, a feature fusion module (TCFF) was added to this network. The two feature maps obtained from Transformer and CNN are utilized to generate Grad-CAM. Subsequently, we use Conditional Random Field (CRF) to further refine the Grad-CAM and adapt Affinity from Attention (AFA), which learn semantic affinity from the Gated Axial Transformer and the Convolutional Neural Network, to produce more accurate pseudo labels.
   • 

3. Crack Detection as a Weakly-Supervised Problem: Towards Achieving Less Annotation-Intensive Crack Detectors, International Conference on Pattern Recognition, 2021. [Paper Link]

   • we formulate the crack detection problem as a weakly-supervised problem and propose a two-branched framework. By combining predictions of a supervised model trained on low quality annotations with predictions based on pixel brightness, our framework is less affected by the annotation quality.
   • Experimental results show that the proposed framework retains high detection accuracy even when provided with low quality annotations.
   • 

4. Investigation of pavement crack detection based on deep learning method using weakly supervised instance segmentation framework, Construction and Building Materials, 2022. [Paper Link]

   • a deep learning method is used to detect cracks based on the weakly supervised instance segmentation (WSIS) framework. A bounding box-level crack image data is preprocessed. Pseudo labels are generated by a region growing algorithm and a GrabCut algorithm.
   • Another important contribution is a new dynamically balanced binary cross-entropy loss function.
   • 

5. Pixel-level tunnel crack segmentation using a weakly supervised annotation approach, Computers in Industry, 2021. [Paper Link]

   • an innovative framework, which combines the weakly supervised learning methods (WSL) and the fully supervised learning methods (FSL), is presented to detect and segment the cracks in the tunnel images.
   • Firstly, a WSL-based segmentation network Crack-CAM is proposed to annotate the collected data instead of using the traditional manual annotation process. By applying the proposed E-Res2Net101 structure and tuning some hyper-parameters, an FSL-based method named DeepLabv3+ is optimized to enhance the segmentation performance. After the crack segmentation, the risk levels of the detected cracks are judged using a new evaluation metric.
   • 

6. Weakly supervised crack segmentation using crack attention networks on concrete structures, Structural Health Monitoring, 2024. [Paper Link]

   • To address this gap, this paper proposes a two-stage weakly supervised learning framework utilizing a novel “crack attention network (CrANET)” with attention mechanism to detect and segment cracks on images with no human annotations in pixel-level labels. This framework classifies concrete surface images into crack or no-cracks and then uses gradient class activation mapping visualization to generate crack segmentation.
   • Professionals and domain experts subsequently evaluate these segmentation results via a human expert validation study
   • 

7. Implementation of surface crack detection method for nuclear fuel pellets by weakly supervised learning, Journal of Nuclear Science and Technology, 2024. [Paper Link]

   • we propose a Weakly Supervised Crack Detection (WSCD) network for surface crack detection of nuclear fuel pellets. The method adopts bounding-box annotations instead of pixel-level annotations, and rapidly generates pixel-level pseudo-labels using the designed Local Fusion Segmentation (LFS) module. Leveraging the Mask-RCNN network as the backbone, the network introduces spatial attention mechanisms to optimize feature extraction networks, enhancing the extraction capability of multi-scale crack features. Lastly, a novel loss function is optimized to address sample imbalance issues and expedite network convergence.
   • Experimental results on the established crack dataset demonstrate that the proposed method improves labeling efficiency by approximately 20 times
   • 

8. A unified approach for weakly supervised crack detection via affine transformation and pseudo label refinement, Scientific Reports, 2025. [Paper Link]

   • This study proposes an Affine Transformation and Pseudo Label Refinement (AT-CAM) method. The methodology comprises three phases: the initial phase employs a geometric enhancement strategy to produce a sequence of enhanced images from the input images, utilizing the Axiom-based Grad- CAM (XGradCAM) algorithm to generate class activation maps for each image, which are subsequently amalgamated into a unified saliency map; in the subsequent phase, the information flow pathways in the subsampling of the convolutional layer are modified by a designated Hook. The information flow in the samples is utilized to invert and eliminate the checkerboard noise produced during integration spatially; in the third stage, a dynamic range compression mechanism is employed to augment the prominence of the cracked areas by compressing the highlighted regions in the saliency map and diminishing the influence of background noise.
   • The experimental results indicate that the method proposed in this study increases segmentation accuracy by 7.2% relative to the original baseline
   • 

9. Weakly-Supervised Surface Crack Segmentation by Generating Pseudo-Labels Using Localization With a Classifier and Thresholding, IEEE Transactions on Intelligent Transportation Systems, 2022. [Paper Link]

   • Our work proposes a weakly supervised approach that leverages a CNN classifier in a novel way to create surface crack pseudo labels. First, we use the classifier to create a rough crack localization map by using its class activation maps and a patch based classification approach and fuse this with a thresholding based approach to segment the mostly darker crack pixels. The classifier assists in suppressing noise from the background regions, which commonly are incorrectly highlighted as cracks by standard thresholding methods. Then, the pseudo labels can be used in an end-to-end approach when training a standard CNN for surface crack segmentation.
   • Our method is shown to yield sufficiently accurate pseudo labels.Those labels, incorporated into segmentation CNN training using multiple recent crack segmentation architectures, achieve comparable performance to fully supervised methods on four popular crack segmentation datasets.
   • 

10. Weakly supervised convolutional neural network for pavement crack segmentation, Intelligent Transportation Infrastructure, 2022. [Paper Link]

    • we proposed a novel Weakly Supervised Learning U-Net (WSL U-Net) for pavement crack segmentation. With the Weakly Supervised Learning (WSL) approach, the training of the network uses weakly labeled images instead of precisely labeled images. The weakly labeled images only need rough labeling, which can significantly alleviate the labor cost and human involvement in image annotation.
    • The dataset cross-validation shows that WSL U-Net outperforms FSL U-Net, suggesting the proposed WSL U-Net is more robust with fewer overfitting concerns and better generalization capability.
    • 
    • 

11. Weakly-supervised structural surface crack detection algorithm based on class activation map and superpixel segmentation, Advances in Bridge Engineering, 2023. [Paper Link]

    • This paper proposes a weakly-supervised structural surface crack detection algorithm that can detect the crack area in an image with low data labeling cost.
    • The algorithm consists of a convolutional neural networks Vgg16-Crack for classification, an improved and optimized class activation map (CAM) algorithm for accurately reflecting the position and distribution of cracks in the image, and a method that combines superpixel segmentation algorithm simple linear iterative clustering (SLIC) with CAM for more accurate semantic segmentation of cracks.
    • 

12. CrackCLIP: Adapting Vision-Language Models for Weakly Supervised Crack Segmentation, Entropy, 2025. [Paper Link]

    • this paper presents CrackCLIP, a novel approach that leverages language prompts to augment the semantic context and employs the Contrastive Language–Image Pre-Training (CLIP) model to enhance weakly supervised crack segmentation.
    • Initially, a gradient-based class activation map is used to generate pixel-level coarse pseudo-labels from a trained crack patch classifier. The estimated coarse pseudo-labels are utilized to fine-tune additional linear adapters, which are integrated into the frozen image encoders of CLIP to adapt the CLIP model to the specialized task of crack segmentation. Moreover, specific textual prompts are crafted for crack characteristics, which are input into the frozen text encoder of CLIP to extract features encapsulating the semantic essence of the cracks. The final crack segmentation is determined by comparing the similarity between text prompt features and visual patch token features.
    • 

13. Weakly Supervised Fatigue Crack Detection in Steel Bridge Girders Using a Proposed Two-Stage Network Training with a Segmentation Refinement Module, Structural Control and Health Monitoring, 2024. [Paper Link]

    • this paper commits to improving the correlation between high-level semantics to low-level appearance.
    • A two-stage training manner with a segmentation refinement module for progressively refining pseudolabels and training the segmentation network was proposed. First, an activation modulation and recalibration scheme was recommended, which leverages a spotlight branch and a compensation branch to locate both the discriminative and less-discriminative object regions.
    • Then, the generated pseudolabels were used as supervision to train the segmentation network in the proposed two-stage manner. In the first stage, the network was pretrained to learn all essential information and provide a basic segmentation performance, aiming to facilitate network convergence in the following training. To develop the inference quality, in the second stage, the pretrained network was further trained recursively with the designed segmentation refinement module to improve the labels using two postprocessing algorithms between each iteration.
    • 

Domain Adaptation

1. Unsupervised domain adaptation for crack detection, Automation in Construction, 2023. [Paper Link]

   • this paper proposes DACrack, an unsupervised domain adaptation framework for crack detection of civil infrastructure. The proposed method performs domain adaptation at the input, feature, and output levels using contrastive mechanisms, adversarial learning, and variational autoencoders.
   • At the input level, foreground enhancement module (FEM) based on the contrast learning mechanism guides the crack detection model to focus on objects with similar structure features to the crack. Then, the feature distribution between different domains is aligned via adversarial learning, namely feature adaptation module (FAM). To further optimize the effect of domain adaptation, variational autoencoder (VAE) is employed to align the clustered manifold structure of cracks on the output space, called output adaptation module (OAM).
   • 

2. Unsupervised domain adaptation-based crack segmentation using transformer network, Journal of Building Engineering, 2023. [Paper Link]

   • we conducted an evaluation of a recent unsupervised domain adaptation model for semantic segmentation that incorporates masked image consistency into DAFormer, a state-of-the-art model with the ability to adapt to various datasets. To assess the model’s performance, we employed three publicly available crack datasets, each containing background and crack classes.
   • Our study has revealed that : (1) SegFormer, a transformer-based model, outperforms ConvNet-based models without utilizing adaptation knowledge, demonstrating superior generalizability to previously unseen data. (2) The unsupervised domain-adaptation model consistently outperforms the source model, resulting in a significant enhancement in the mean intersection over union of SegFormer’s source-only approach by a remarkable 10% to 22%
   • 

3. Self-training with Bayesian neural networks and spatial priors for unsupervised domain adaptation in crack segmentation, Computer-Aided Civil and Infrastructure Engineering, 2024. [Paper Link]

   • This study proposes a novel self-training framework for unsupervised domain adaptation in the segmentation of concrete wall cracks using accumulated crack data. The proposed method incorporates Bayesian neural networks for uncertainty estimation of pseudo-labels, and spatial priors of cracks for screening noisy labels (Since cracks have unique shapes, calculating geometric features allows us to determine the validity of pseudo-labels. The key parameter characterizing cracks is area/perimeter2 due to their elongated shape.).
   • Furthermore, the integration of Stable Diffusion for few-shot image generation enhances domain adaptation performance by 0.0332. The proposed framework enables high-precision crack segmentation with as few as 100 target images, which can be easily obtained at the site, reducing the cost of model deployment in infrastructure maintenance.
   • 
   • 

4. Generative adversarial network based on domain adaptation for crack segmentation in shadow environments, Computer-Aided Civil and Infrastructure Engineering, 2025. [Paper Link]

   • this study proposes a two-stage domain adaptation framework called GAN-DANet for crack segmentation in shadowed environments. In the first stage, CrackGAN uses adversarial learning to merge features from shadow-free and shadowed datasets, creating a new dataset with more domain-invariant features. In the second stage, the CrackSeg network innovatively integrates enhanced Laplacian filtering (ELF) into high-resolution net to enhance crack edges and texture features while filtering out shadow information.
   • CrackGAN addresses domain shift by generating a new dataset with domain-invariant features, avoiding direct feature alignment between source and target domains. The ELF module in CrackSeg effectively enhances crack features and suppresses shadow interference, improving the segmentation model's robustness in shadowed environments.
   • 

5. CrackUDA: Incremental Unsupervised Domain Adaptation for Improved Crack Segmentation in Civil Structures, International Conference on Pattern Recognition, 2024. [Paper Link]

   • we propose a novel deep network that employs incremental training with unsupervised domain adaptation (UDA) using adversarial learning, without a significant drop in accuracy in the source domain. Our approach leverages an encoder-decoder architecture, consisting of both domain-invariant and domain-specific parameters. The encoder learns shared crack features across all domains, ensuring robustness to domain variations. Simultaneously, the decoder’s domain-specific parameters capture domain-specific features unique to each domain.
   • Furthermore, we introduce BuildCrack, a new crack dataset comparable to sub-datasets of the well-established CrackSeg9K dataset in terms of image count and crack percentage.
   • 

6. Shape-Consistent One-Shot Unsupervised Domain Adaptation for Rail Surface Defect Segmentation, IEEE Transactions on Industrial Informatics, 2023. [Paper Link]

   • Conventional deep learning models show limited generalization in scenes with distribution differences. To address this problem, we propose a novel one-shot unsupervised domain adaptation framework. Specifically, we introduce a shape-consistent style transfer module that performs pixel-level distribution alignment between the training and test images. Based on the one-shot test image, the training image is reconstructed to have the same appearance as the test image.
   • Meanwhile, we employ a multitask learning strategy to prevent content distortion of the reconstructed images. To improve the robustness of the model to distribution differences, we design an edge-aware defect segmentation model and train the model using the reconstructed training images.
   • 

7. Multi-source Domain Adaptation for Unsupervised Road Defect Segmentation, IEEE International Conference on Robotics and Automation (ICRA), 2023. [Paper Link]

   • we propose a novel multi-source domain adaptation method to boost the performance of road defect segmentation on an unlabelled dataset. The proposed method generates multi-source ensembled labels using transferred information from models trained with multiple labelled source domains, which are utilised as supervisory signals for the unlabelled target domain.
   • Furthermore, to reduce the domain gap between each source domain and a target domain, these domains are re-aligned with outlier repositioning to improve the defect segmentation performance. We demonstrate the effectiveness of our proposed method on Cracktree200, CRACK500, CFD, and Crack360 datasets.
   • 

8. Industrial UAV-Based Unsupervised Domain Adaptive Crack Recognitions: From Database Towards Real-Site Infrastructural Inspections, IEEE Transactions on Industrial Electronics, 2022. [Paper Link]

   • we propose a robust UDA learning strategy termed Crack-DA to increase the generalization capacity of the model in unseen test circumstances.
   • More specifically, we first propose leveraging the self-supervised depth information to help the learning of semantics. And then using the edge information to suppress nonedge background objects and noises. We also use the data augmentation-based consistency. More importantly, we use the disparity in depth to evaluate the domain gap in semantics and explicitly consider the domain gap in network optimization. A database consisting of 11 298 crack images with detailed pixel-level labels for network training in domain adaptations is established
   • 

9. Deep convolutional transfer learning-based structural damage detection with domain adaptation, *Applied Intelligence *, 2022. [Paper Link]

   • a novel structural damage detection method is proposed by using deep convolutional transfer learning. In the method, one-dimensional deep convolutional neural network and two-dimensional deep convolutional neural network are combined to mine more fine-grained features with spatiotemporal characteristic from raw vibration data.
   • And, a novel domain adaptation technology, combining multikernel maximum mean discrepancy and local maximum mean discrepancy, is developed to align the distribution of global domains and relevant subdomains among different domains, which could mine more fine-grained features for each category and improve the transfer performance. Transfer experiments on two different structures are implemented to verify the effectiveness of the proposed method. Furthermore, a new solution is found by taking advantage of the damage knowledge learnt from other structure to implement damage detection when very small damage samples are available.
   • 

10. Cross-dataset semantic segmentation for composite crack detection using unsupervised transfer learning, Composite Structures, 2025. [Paper Link]

    • This study proposes an unsupervised framework that integrates hybrid feature- and instance-based transfer learning techniques, investigating transfer directions and data reconstruction. Specifically, a Cycle Generative Adversarial Network is employed to align features without annotating target data. Additionally, a U-Net architecture enhanced with Squeeze-and-Excitation attention mechanisms—establishing weight relationships through channel-wise feature mapping—is utilized to improve crack feature extraction and segmentation accuracy.
    • 

11. Self-training method for structural crack detection using image blending-based domain mixing and mutual learning, Automation in Construction, 2025. [Paper Link]

    • This paper addresses the issues by introducing a robust self-training domain adaptive segmentation (STDASeg) pipeline. STDASeg incorporates an image blending-based domain mixing module to minimize domain discrepancies.
    • Additionally, STDASeg involves a two-stage self-training framework characterized by the mutual learning scheme between Convolutional Neural Networks and Transformers, effectively learning domain invariant features from the two domains.
    • 

Few shot

1. A Few-Shot Attention Recurrent Residual U-Net for Crack Segmentation, ISVC, 2023. [Paper Link]

   • Recent studies indicate that deep learning plays a crucial role in the automated visual inspection of road infrastructures. However, current learning schemes are static, implying no dynamic adaptation to users’ feedback.
   • To address this drawback, we present a few-shot learning paradigm for the automated segmentation of road cracks, which is based on a U-Net architecture with recurrent residual and attention modules (R2AU-Net). The retraining strategy dynamically fine-tunes the weights of the U-Net as a few new rectified samples are being fed into the classifier.
   • 

2. CrackNex: a Few-shot Low-light Crack Segmentation Model Based on Retinex Theory for UAV Inspections, ICRA, 2024. [Paper Link][Code Link]

   • Crack segmentation under such conditions is challenging due to the poor contrast between cracks and their surroundings. However, most deep learning methods are designed for well-illuminated crack images and hence their performance drops dramatically in low-light scenes. In addition, conventional approaches require many annotated low-light crack images which is time-consuming.
   • we address these challenges by proposing CrackNex, a framework that utilizes reflectance information based on Retinex Theory to learn a unified illumination-invariant representation. Furthermore, we utilize few-shot segmentation to solve the inefficient training data problem. In CrackNex, both a support prototype and a reflectance prototype are extracted from the support set. Then, a prototype fusion module is designed to integrate the features from both prototypes. CrackNex outperforms the SOTA methods on multiple datasets.
   • Additionally, we present the first benchmark dataset, LCSD, for low-light crack segmentation
   • 

3. Cycle-consistency-constrained few-shot learning framework for universal multi-type structural damage segmentation, Structural Health Monitoring, 2024. [Paper Link]

   • This study proposes a novel cycle-consistency-constrained few-shot segmentation framework tailored for multi-type structural damage recognition. A cycle-consistency-constrained prototype learning paradigm is constructed to enhance the adequate utilization of limited pixel-level annotations, which is leveraged by establishing a bidirectional mutual supervision mechanism between support and query sets.
   • Subsequently, a non-parametric similarity-guided optimization module is incorporated into the high-level latent feature space of image embedding. This module induces a similarity-driven contrast learning process for each pixel of feature maps and learns universal prototypes that condense the abstract semantic context of foreground (i.e., multi-type damage) and background.
   • Furthermore, a synthetic loss function, which comprises mutually supervised segmentation dice loss, metric loss, and contrastive loss, is designed to ensure the bidirectional pixel-level segmentation accuracy, intra-class compactness, and inter-class separability of learned prototypes for multi-type damage
   • 

4. A Suspected Defect Screening-Guided Lightweight Network for Few-Shot Aviation Steel Tube Surface Defect Segmentation, IEEE Sensors Journal, 2024. [Paper Link]

   • In this work, a lightweight suspected defect screening network (SDSNet) based on transfer learning is proposed to filter out redundant nondefect samples. Besides, to overcome the comprehensive sparsity of AST defect samples, we introduce a granularity-transfer few-shot defect segmentation (FSDS) based on meta-learning.
   • Subsequently, we propose a lightweight feature-aware segmentation network (FASNet) further to segment the suspected defect sample pixel-wise. Specifically, a defect-aware module (DAM) is employed to activate spatial and channel responses of defect regions, and a lightweight multiscale aggregation decoder (MAD) is used to capture context information at different feature scales.
   • In addition, to evaluate the effectiveness of the proposed pipeline and overcome the practical challenge of AST defect segmentation, a dedicated database is constructed.
   • 
   • 

5. Adaptive Cross-Scenario Few-Shot Learning Framework for Structural Damage Detection in Civil Infrastructure, Journal of Construction Engineering and Management, 2023. [Paper Link]

   • this paper proposes a novel framework for structural damage detection with large scope of cross-task learning capability that incorporates Bayesian estimation and variational inference into the deep learning backbones and Bayesian weight function into the outer loop process of metalearning.
   • Experimental results demonstrate the superiority of this method for both structural damage image classification and structural damage semantic segmentation. Compared with existing frameworks, the proposed method can alleviate the negative influence of domain bias and reduce computation time and costs due to sample labeling.
   • 

6. Task-aware meta-learning paradigm for universal structural damage segmentation using limited images, Engineering Structures, 2023. [Paper Link]

   • This study proposes a task-aware meta-learning paradigm using limited images for universal structural damage segmentation. First, an interpretable task generation strategy instead of random sampling is designed based on feature density clustering, and a synthetical metric of Jaccard distance and Euclidean distance is established to measure the feature similarity and discover the class separability in the high-level feature space.
   • Second, a dual-stage optimization framework is built based on Model-Agnostic Meta-Learning (MAML), comprising an internal optimization of the inner semantic segmentation model and an external optimization of the meta-learning machine.
   • Third, core samples around the cluster center are selected to form a query pool and evaluate the task-significance scores of different tasks within a meta-batch, which are utilized in the external optimization to control the orientation of gradient updates towards more significant tasks.
   • Finally, a multi-type structural damage dataset, including concrete crack, steel fatigue crack, concrete spalling, cable corrosion, and cable clamp slipping, is utilized to verify the effectiveness and necessity.
   • 

7. Multi-Type Structural Damage Image Segmentation via Dual-Stage Optimization-Based Few-Shot Learning, Smart Cities, 2024. [Paper Link]

   • this study proposes a dual-stage optimization-based few-shot learning segmentation method using only a few images with supervised information for multi-type structural damage recognition. A dual-stage optimization paradigm is established encompassing an internal network optimization based on meta-task and an external meta-learning machine optimization based on meta-batch.
   • The underlying image features pertinent to various structural damage types are learned as prior knowledge to expedite adaptability across diverse damage categories via only a few samples. Furthermore, a mathematical framework of optimization-based few-shot learning is formulated to intuitively express the perception mechanism.
   • 

8. Recovering compressed images for automatic crack segmentation using generative models, Mechanical Systems and Signal Processing, 2021. [Paper Link]

   • we present a new approach of CS that replaces the sparsity regularization with a generative model that is able to effectively capture a low dimension representation of targeted images. We develop a recovery framework for automatic crack segmentation of compressed crack images based on this new CS method.
   • We demonstrate the remarkable performance of our method that takes advantage of the strong capability of generative models to capture the necessary features required in the crack segmentation task even the backgrounds of the generated images are not well reconstructed.
   • 

9. Few-shot learning for structural health diagnosis of civil infrastructure, Advanced Engineering Informatics, 2024. [Paper Link]

   • This article systematically summarizes recent advances in FSL algorithms and the corresponding applications in SHD for civil infrastructure. A unified mathematical framework of FSL is formulated, and an FSL taxonomy is summarized according to intrinsic learning mechanisms and implementation principles, including metric learning-based, optimization-based, transfer learning-based, and generative model-based methods. Various applications of SHD for civil infrastructure under real-world scenarios are reviewed, including remote sensing monitoring, structural damage recognition, post-disaster safety evaluation, and construction risk assessment.
   • 

Unsupervised

1. Self-Supervised Structure Learning for Crack Detection Based on Cycle-Consistent Generative Adversarial Networks, Journal of Computing in Civil Engineering, 2020. [Paper Link]

   • This paper proposes a self-supervised structure learning network which can be trained without using paired data, even without using ground truths (GTs); this is achieved by training an additional reverse network to translate the output back to the input simultaneously.
   • First, a labor-free structure library is prepared and set as the target domain for structure learning. Then a dual network is built with two generative adversarial networks (GANs); one is trained to translate a crack image patch (X) to a structural patch (Y), and the other is trained to translate Y back to X, simultaneously.
   • The experiments demonstrated that with such settings, the network can be trained to translate a crack image to the GT-like image with a similar structure pattern, and it can be used for crack detection. The proposed approach was validated on four crack data sets and achieved comparable performance to that of state-of-the-art supervised approaches.
   • 

2. Unsupervised Pixel-level Crack Detection Based on Generative Adversarial Network, ICMSSP, 2020. [Paper Link]

   • we present an unsupervised method for learning mapping to translate crack images to binary images based on generative adversarial network. We introduce the cyclic consistent loss to increase accuracy of crack localization.
   • Eight residual blocks connected convolutional neural network for feature extraction is used as generator and a 5-layer fully convolutional network is used as discriminator. We analyze the proposed framework and provide qualitative and quantitative comparison. The experimental results show that the proposed method achieves a better performance than several existing methods.
   • 

3. UP-CrackNet: Unsupervised Pixel-Wise Road Crack Detection via Adversarial Image Restoration, IEEE Transactions on Intelligent Transportation Systems, 2024. [Paper Link]

   • we propose an unsupervised pixel-wise road crack detection network, known as UP-CrackNet. Our approach first generates multi-scale square masks and randomly selects them to corrupt undamaged road images by removing certain regions. Subsequently, a generative adversarial network is trained to restore the corrupted regions by leveraging the semantic context learned from surrounding uncorrupted regions.
   • During the testing phase, an error map is generated by calculating the difference between the input and restored images, which allows for pixel-wise crack detection. Our comprehensive experimental results demonstrate that UP-CrackNet outperforms other general-purpose unsupervised anomaly detection algorithms
   • 

4. Anomaly detection of defects on concrete structures with the convolutional autoencoder, Advanced Engineering Informatics, 2020. [Paper Link]

   • A convolutional autoencoder was trained as a reconstruction-based model, with the defect-free images, to rapidly and reliably detect defects from the large volume of image datasets. This training process was in the unsupervised mode, with no label needed, thereby requiring no prior knowledge and saving an enormous amount of time for label preparation.
   • The built anomaly detector favors minimizing the reconstruction errors of defect-free images, which renders high reconstruction errors of defects, in turn, detecting the location of defects. The assessment shows that the proposed anomaly detection technique is robust and adaptable to defects on wide ranges of scales.
   • 

5. Anomaly detection of cracks in synthetic masonry arch bridge point clouds using fast point feature histograms and PatchCore, Automation in Construction, 2024. [Paper Link]

   • Previous research has focused on detecting surface cracks from images. This paper develops an alternative approach where cracks are identified from point clouds via geometric distortions.
   • An image-based anomaly detection method called PatchCore is customized for 3D applications for this purpose. First, Fast Point Feature Histograms (FPFH) are used to extract geometric features. Then PatchCore is applied on synthetic point clouds with crack labels, generated using 3D finite element modelling (FEM) and graphical modelling.
   • 

Foundation Models

1. Fine-tuning vision foundation model for crack segmentation in civil infrastructures, Construction and Building Materials, 2024. [Paper Link]

   • In this work, a vision foundation model is introduced for crack segmentation. Two parameter-efficient fine-tuning methods, adapter and low-rank adaptation, are adopted to fine-tune the foundation model in semantic segmentation: the Segment Anything Model (SAM).
   • The fine-tuned CrackSAM shows excellent performance on different scenes and materials. To test the zero-shot performance of the proposed method, two unique datasets related to road and exterior wall cracks are collected, annotated and open-sourced, for a total of 810 images.
   • 

2. Segment anything model-based crack segmentation using low-rank adaption fine-tuning, Structural Health Monitoring, 2024. [Paper Link]

   • This study introduces a novel approach that fine-tunes SAM specifically for crack segmentation by incorporating low-rank adaptation (LoRA). This method involves adding a dedicated crack segmentation head to SAM, enabling automatic crack segmentation. Additionally, the application of LoRA technology facilitates the readjustment of SAM’s features without incurring the substantial costs typically associated with fine-tuning entire networks.
   • A comparative analysis with current leading crack segmentation models demonstrated a significant increase in accuracy across eight different crack datasets.
   • 

3. Semi-supervised crack detection using segment anything model and deep transfer learning, Automation in Construction, 2025. [Paper Link]

   • This paper proposes a semi-supervised instance segmentation method for road distress detection based on deep transfer learning. The interactive segmentation method utilizing SAM are used to enhance the production efficiency of segmentation datasets. The DCNv3 and lightweight segmentation heads are strategically designed to offset potential speed losses. The deep transfer learning method fine-tunes the pre-trained models, enhancing their competency for new tasks.
   • 

4. Segment Any Crack: Deep Semantic Segmentation Adaptation for Crack Detection, Arxiv, 2025. [Paper Link]

   • This study introduces an efficient selective fine-tuning strategy, focusing on tuning normalization components, to enhance the adaptability of segmentation models for crack detection. The proposed method is applied to the Segment Anything Model (SAM) and five well-established segmentation models. Experimental results demonstrate that selective fine- tuning of only normalization parameters outperforms full fine-tuning and other common fine-tuning techniques in both performance and computational efficiency, while improving generalization.
   • The proposed approach yields a SAM-based model, Segment Any Crack (SAC), achieving a 61.22% F1-score and 44.13% IoU on the OmniCrack30k benchmark dataset, along with the highest performance across three zero-shot datasets and the lowest standard deviation
   • 

5. Sam-based instance segmentation models for the automation of structural damage detection, Advanced Engineering Informatics, 2024. [Paper Link]

   • Previous studies mainly focused on concrete structures and pavements, neglecting masonry defects and lacking publicly available datasets. In this paper, we address these gaps by introducing the “MCrack1300” dataset, annotated for bricks, broken bricks, and cracks, targeting instance segmentation.
   • We propose two novel, automatically executable methods based on the latest visual large-scale model, the prompt-based Segment Anything Model (SAM). We fine-tune SAM’s encoder using Low-Rank Adaptation (LoRA). The first method connects SAM’s encoder to other decoders directly, while the second uses a learnable self-generating prompter. We modify the feature extractor for seamless integration of these methods with SAM’s encoder. Both methods outperform the state-of-the-art models, improving benchmark results approximately 3 % across all classes and around 6 % specifically for cracks
   • 
   • 

6. CrackESS: A Self-Prompting Crack Segmentation System for Edge Devices, Arxiv, 2024. [Paper Link]

   • Recent CNN-based and SAM-based approaches have demon- strated excellent performance in crack segmentation, but their high computational demands limit their applicability on edge devices.
   • This paper introduces CrackESS, a novel system for detecting and segmenting concrete cracks. The approach first utilizes a YOLOv8 model for self-prompting and a LoRA-based fine-tuned SAM model for crack segmentation, followed by refining the segmentation masks through the proposed Crack Mask Refinement Module (CMRM).
   • 

Citation

If you find this repository helpful, please consider giving a star and citing:

@article{zhang2025deep,
  title={Deep Learning for Crack Detection: A Review of Learning Paradigms, Generalizability, and Datasets},
  author={Zhang, Xinan and Wang, Haolin and Hsieh, Yung-An and Yang, Zhongyu and Yezzi, Anthony and Tsai, Yi-Chang},
  journal={arXiv preprint arXiv:2508.10256},
  year={2025}
}