A Novel Approach to Data Augmentation for Pavement Distress Segmentation

Mazzini, Davide; Napoletano, Paolo; Piccoli, Flavio; Schettini, Raimondo

doi:10.1016/j.compind.2020.103225

Cited by 40 publications

(19 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The disparity of MDA to conventional DA is that MDA utilizes a large number of different data augmentation methods. There are two types [45] of MDA, offline and online. Offline means editing and storing data on the disk, and online means on-the-fly augmentation.…”

Section: Methodsmentioning

confidence: 99%

COVID-19 classification by CCSHNet with deep fusion using transfer learning and discriminant correlation analysis

et al. 2021

View full text Add to dashboard Cite

Highlights We proposed a novel (L, 2) transfer feature learning (L2TFL) approach. L2TFL can elucidate the optimal layers to be removed prior to selection. We developed a novel selection algorithm of pretrained network for fusion approach. SAPNF can determine the best two pretrained models for fusion. We introduced a deep CCT fusion discriminant correlation analysis fusion method.

show abstract

Section: Methodsmentioning

confidence: 99%

COVID-19 classification by CCSHNet with deep fusion using transfer learning and discriminant correlation analysis

et al. 2021

View full text Add to dashboard Cite

show abstract

“…This is particularly useful when available labeled data is scarce. Motivated by the difficulty to obtain annotated images for crack detection, offline data augmentation has been used too (Mazzini et al, 2020;Kanaeva and Ivanova, 2021). However, these approaches rely on inaccurate annotations, carrying the bias introduced by manual annotation.…”

Section: Related Literaturementioning

confidence: 99%

Syncrack: Improving Pavement and Concrete Crack Detection through Synthetic Data Generation

Rill-García

Dokladalova

Dokládal

2022

Proceedings of the 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Application

View full text Add to dashboard Cite

In crack detection, pixel-accurate predictions are necessary to measure the width -an important indicator of the severity of a crack. However, manual annotation of images to train supervised models is a hard and time-consuming task. Because of this, 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. By using this, first we do a robustness study of the impact of training with inaccurate labels. This study quantifies the detrimental effect of inaccurate annotations in the final prediction scores. Afterwards, we propose to use Syncrack to avoid this detrimental effect in a real-life context. For this, we show the advantages of using Syncrack generated images with accurate annotations for crack detection on real road images. Since supervised scores are biased by the inaccuracy of annotations, we propose a set of unsupervised metrics to evaluate the segmentation quality in terms of crack width. a https://orcid.

show abstract

“…Furthermore, loss functions based on VGG-extracted feature maps have been used for texture generation. Particularly, to reduce the impact of limited amounts of available annotated data, [47] used a semantic texture generation approach for data augmentation. This improved the model's performance in real images.…”

Section: U-vgg19mentioning

confidence: 99%