Road crack segmentation using an attention residual U-Net with generative adversarial

This paper proposed an end-to-end road crack segmentation model based on attention mechanism and deep FCN with generative adversarial learning. We create a segmentation network by introducing a visual attention mechanism and residual module to a fully convolutional network(FCN) to capture richer loc...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Xing Hu, Minghui Yao, Dawei Zhang
Formato:	article
Lenguaje:	EN
Publicado:	AIMS Press 2021
Materias:	road crack segmentation attention residual u-net adversarial learning Biotechnology TP248.13-248.65 Mathematics QA1-939
Acceso en línea:	https://doaj.org/article/ccc0569333544ed7a9cc36fc49156950
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:ccc0569333544ed7a9cc36fc49156950
record_format	dspace
spelling	oai:doaj.org-article:ccc0569333544ed7a9cc36fc491569502021-11-29T06:31:36ZRoad crack segmentation using an attention residual U-Net with generative adversarial10.3934/mbe.20214731551-0018https://doaj.org/article/ccc0569333544ed7a9cc36fc491569502021-11-01T00:00:00Zhttps://www.aimspress.com/article/doi/10.3934/mbe.2021473?viewType=HTMLhttps://doaj.org/toc/1551-0018This paper proposed an end-to-end road crack segmentation model based on attention mechanism and deep FCN with generative adversarial learning. We create a segmentation network by introducing a visual attention mechanism and residual module to a fully convolutional network(FCN) to capture richer local features and more global semantic features and get a better segment result. Besides, we use an adversarial network consisting of convolutional layers as a discrimination network. The main contributions of this work are as follows: 1) We introduce a CNN model as a discriminate network to realize adversarial learning to guide the training of the segmentation network, which is trained in a min-max way: the discrimination network is trained by maximizing the loss function, while the segmentation network is trained with the only gradient passed by the discrimination network and aim at minimizing the loss function, and finally an optimal segmentation network is obtained; 2) We add the residual modular and the visual attention mechanism to U-Net, which makes the segmentation results more robust, refined and smooth; 3) Extensive experiments are conducted on three public road crack datasets to evaluate the performance of our proposed model. Qualitative and quantitative comparisons between the proposed method and the state-of-the-art methods show that the proposed method outperforms or is comparable to the state-of-the-art methods in both F1 score and precision. In particular, compared with U-Net, the mIoU of our proposed method is increased about 3%~17% compared with the three public datasets.Xing HuMinghui Yao Dawei ZhangAIMS Pressarticleroad crack segmentationattention residual u-netadversarial learningBiotechnologyTP248.13-248.65MathematicsQA1-939ENMathematical Biosciences and Engineering, Vol 18, Iss 6, Pp 9669-9684 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	road crack segmentation attention residual u-net adversarial learning Biotechnology TP248.13-248.65 Mathematics QA1-939
spellingShingle	road crack segmentation attention residual u-net adversarial learning Biotechnology TP248.13-248.65 Mathematics QA1-939 Xing Hu Minghui Yao Dawei Zhang Road crack segmentation using an attention residual U-Net with generative adversarial
description	This paper proposed an end-to-end road crack segmentation model based on attention mechanism and deep FCN with generative adversarial learning. We create a segmentation network by introducing a visual attention mechanism and residual module to a fully convolutional network(FCN) to capture richer local features and more global semantic features and get a better segment result. Besides, we use an adversarial network consisting of convolutional layers as a discrimination network. The main contributions of this work are as follows: 1) We introduce a CNN model as a discriminate network to realize adversarial learning to guide the training of the segmentation network, which is trained in a min-max way: the discrimination network is trained by maximizing the loss function, while the segmentation network is trained with the only gradient passed by the discrimination network and aim at minimizing the loss function, and finally an optimal segmentation network is obtained; 2) We add the residual modular and the visual attention mechanism to U-Net, which makes the segmentation results more robust, refined and smooth; 3) Extensive experiments are conducted on three public road crack datasets to evaluate the performance of our proposed model. Qualitative and quantitative comparisons between the proposed method and the state-of-the-art methods show that the proposed method outperforms or is comparable to the state-of-the-art methods in both F1 score and precision. In particular, compared with U-Net, the mIoU of our proposed method is increased about 3%~17% compared with the three public datasets.
format	article
author	Xing Hu Minghui Yao Dawei Zhang
author_facet	Xing Hu Minghui Yao Dawei Zhang
author_sort	Xing Hu
title	Road crack segmentation using an attention residual U-Net with generative adversarial
title_short	Road crack segmentation using an attention residual U-Net with generative adversarial
title_full	Road crack segmentation using an attention residual U-Net with generative adversarial
title_fullStr	Road crack segmentation using an attention residual U-Net with generative adversarial
title_full_unstemmed	Road crack segmentation using an attention residual U-Net with generative adversarial
title_sort	road crack segmentation using an attention residual u-net with generative adversarial
publisher	AIMS Press
publishDate	2021
url	https://doaj.org/article/ccc0569333544ed7a9cc36fc49156950
work_keys_str_mv	AT xinghu roadcracksegmentationusinganattentionresidualunetwithgenerativeadversarial AT minghuiyao roadcracksegmentationusinganattentionresidualunetwithgenerativeadversarial AT daweizhang roadcracksegmentationusinganattentionresidualunetwithgenerativeadversarial
_version_	1718407551612092416

Road crack segmentation using an attention residual U-Net with generative adversarial

Ejemplares similares