Deployment Conscious Automatic Surface Crack Detection

Inoue, Yoshio; Nagayoshi, Hiroto

doi:10.1109/wacv.2019.00078

Cited by 16 publications

(22 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…More specifically, the Macro F1-score 1 is evaluated because it weighs all images equally toward the overall score, unlike the Micro F1-score which weighs images with thicker cracks more. Also, pixel tolerances, typically introduced to absorb the annotation inaccuracies [1], [13], are not adopted because it ambiguates the effect of low quality annotations.…”

Section: Methodsmentioning

confidence: 99%

“…Examples include the work of Zhang et al, in which a shallow CNN was used for prediction [12], and work by Fan et al, in which the idea of structured prediction is introduced to force the model to learn the relationship between neighboring pixels [1]. Inoue et al proposed the Multiple Instance Learning architecture (MIL) which resembles test time augmentation (except it is also applied during training time as well) to increase the rotational robustness of the network [13]. Other works use deeper CNNs with encoderdecoder architecture with skip connections, to better fuse information from multiple scales [5], [6].…”

Section: A Crack Detection Approachesmentioning

confidence: 99%

“…To assess the versatility of our proposed framework, it was tested with four semantic segmentation model architectures. First, DeepCrack by Liu et al [6] 2 and the model proposed by Inoue et al [13], which are both architectures designed specifically for crack detection, are selected to assess the effectiveness of the proposed framework for task-specific models. In addition, DeepLab V3+ [23] 3 , which is designed for general semantic segmentation tasks, is selected to evaluate the proposed framework on general task models.…”

Section: B Macro Branch Modelsmentioning

confidence: 99%

See 2 more Smart Citations

Crack Detection as a Weakly-Supervised Problem: Towards Achieving Less Annotation-Intensive Crack Detectors

Inoue

Nagayoshi

2020

Preprint

Self Cite

View full text Add to dashboard Cite

Automatic crack detection is a critical task that has the potential to drastically reduce labor-intensive building and road inspections currently being done manually. Recent studies in this field have significantly improved the detection accuracy. However, the methods often heavily rely on costly annotation processes. In addition, to handle a wide variety of target domains, new batches of annotations are usually required for each new environment. This makes the data annotation cost a significant bottleneck when deploying crack detection systems in real life. To resolve this issue, 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. Implementation of the proposed framework is publicly available at https://github. com/hitachi-rd-cv/weakly-sup-crackdet.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: A Crack Detection Approachesmentioning

confidence: 99%

Section: B Macro Branch Modelsmentioning

confidence: 99%

See 1 more Smart Citation

Crack Detection as a Weakly-Supervised Problem: Towards Achieving Less Annotation-Intensive Crack Detectors

Inoue

Nagayoshi

2020

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Other tasks with which the state of structural cracks can be assessed may include quantification analysis such as determining the rotation or providing measurements of cracks. Rotation of cracks may be important because maintenance measures may only need to be applied if the crack has a certain orientation [39]. However, only a few DL based works cover the orientation prediction of road cracks [39], [40].…”

Section: Other Tasksmentioning

confidence: 99%

What's Cracking? A Review and Analysis of Deep Learning Methods for Structural Crack Segmentation, Detection and Quantification

König¹,

Jenkins²,

Mannion³

et al. 2022

Preprint

View full text Add to dashboard Cite

Surface cracks are a very common indicator of potential structural faults. Their early detection and monitoring is an important factor in structural health monitoring. Left untreated, they can grow in size over time and require expensive repairs or maintenance. With recent advances in computer vision and deep learning algorithms, the automatic detection and segmentation of cracks for this monitoring process have become a major topic of interest. This review aims to give researchers an overview of the published work within the field of crack analysis algorithms that make use of deep learning. It outlines the various tasks that are solved through applying computer vision algorithms to surface cracks in a structural health monitoring setting and also provides in-depth reviews of recent fully, semi and unsupervised approaches that perform crack classification, detection, segmentation and quantification. Additionally, this review also highlights popular datasets used for cracks and the metrics that are used to evaluate the performance of those algorithms. Finally, potential research gaps are outlined and further research directions are provided.

show abstract

“…DRIVE has 40 color fundus photographs. For training and test sets, we split AigleRN equally into 3 folds for cross validation, split CFD into 71 training images and 46 testing images as the setting of [9,11,23], and split DRIVE into 20 training images and 20 testing images as the setting of [18,19].…”

Section: Experiments 31 Experimental Setupmentioning

confidence: 99%

Crack Detection and Refinement Via Deep Reinforcement Learning

Park

Chen

2021

2021 IEEE International Conference on Image Processing (ICIP)

View full text Add to dashboard Cite

Detecting small cracks in concrete is difficult due to the complexity and thinness of cracking patterns, which requires the development of refined vision-based segmentation algorithms that can accurately characterize the details of crack defects. While existing methods are good at generally outlining cracks, due to inherent differences in shape distributions between common objects and cracks, their predictions often have disconnected segments and inaccuracy along boundaries. To this end, we develop a refinement framework using reinforcement learning (RL) that can better recognize details specific to cracks. Our method uses an RL agent to iteratively improve per-pixel crack predictions of a general segmentation model. We find that in addition to connecting gaps in predictions, the RL agent is also able to detect cracks that are missed in the original predictions. It does so by using the originally detected regions as crack priors to branch out from. Refining outputs of a commonly used per-pixel segmentation model, our method outperforms the current state-of-theart approaches for crack segmentation. Our experiments also demonstrate that our method generalizes well to a similar task of vessel segmentation.

show abstract

Deployment Conscious Automatic Surface Crack Detection

Cited by 16 publications

References 13 publications

Crack Detection as a Weakly-Supervised Problem: Towards Achieving Less Annotation-Intensive Crack Detectors

Crack Detection as a Weakly-Supervised Problem: Towards Achieving Less Annotation-Intensive Crack Detectors

What's Cracking? A Review and Analysis of Deep Learning Methods for Structural Crack Segmentation, Detection and Quantification

Crack Detection and Refinement Via Deep Reinforcement Learning

Contact Info

Product

Resources

About