Crack detection based on ResNet with spatial attention

被引：6

作者：

Yang, Qiaoning ^{[1
]}

Jiang, Si ^{[1
]}

Chen, Juan ^{[1
]}

Lin, Weiguo ^{[1
]}

机构：

[1] Beijing Univ Chem Technol, Coll Informat Sci & Technol, Beijing 100029, Peoples R China

来源：

COMPUTERS AND CONCRETE | 2020年 / 26卷 / 05期

关键词：

crack detection; attention mechanism; deep convolution neural network; SYSTEM;

D O I：

10.12989/cac.2020.26.5.411

中图分类号：

TP39 [计算机的应用];

学科分类号：

081203 ; 0835 ;

摘要：

Deep Convolution neural network (DCNN) has been widely used in the healthy maintenance of civil infrastructure. Using DCNN to improve crack detection performance has attracted many researchers' attention. In this paper, a light-weight spatial attention network module is proposed to strengthen the representation capability of ResNet and improve the crack detection performance. It utilizes attention mechanism to strengthen the interested objects in global receptive field of ResNet convolution layers. Global average spatial information over all channels are used to construct an attention scalar. The scalar is combined with adaptive weighted sigmoid function to activate the output of each channel's feature maps. Salient objects in feature maps are refined by the attention scalar. The proposed spatial attention module is stacked in ResNet50 to detect crack. Experiments results show that the proposed module can got significant performance improvement in crack detection.

引用

页码：411 / 420

页数：10

共 37 条

[21]

Lee D, 2019, INT J AERONAUT SPACE, V20, P287

[22] Motion Guided Attention for Video Salient Object Detection [J].

Li, Haofeng ;

Chen, Guanqi ;

Li, Guanbin ;

Yu, Yizhou .

2019 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2019), 2019, :7273-7282

[23] The role of public-private partnership in constructing the smart transportation city: a case of the bike sharing platform [J].

Li, Lin ;

Park, Philip ;

Yang, Sung-Byung .

ASIA PACIFIC JOURNAL OF TOURISM RESEARCH, 2021, 26 (04) :428-439

[24] Real-time comprehensive image processing system for detecting concrete bridges crack [J].

Lin, Weiguo ;

Sun, Yichao ;

Yang, Qiaoning ;

Lin, Yaru .

COMPUTERS AND CONCRETE, 2019, 23 (06) :445-457

[25]

Mnih V., 2014, ADV NEUR IN

[26] The dynamic representation of scenes [J].

Rensink, RA .

VISUAL COGNITION, 2000, 7 (1-3) :17-42

[27] Morphological segmentation based on edge detection-II for automatic concrete crack measurement [J].

Su, Tung-Ching ;

Yang, Ming-Der .

COMPUTERS AND CONCRETE, 2018, 21 (06) :727-739

[28]

Szegedy Christian, 2015, IEEE C COMPUTER VIS, P1, DOI [10.1109/cvpr.2015.7298594, DOI 10.1109/CVPR.2015.7298594, 10.1109/CVPR.2015.7298594]

[29] Residual Attention Network for Image Classification [J].

Wang, Fei ;

Jiang, Mengqing ;

Qian, Chen ;

Yang, Shuo ;

Li, Cheng ;

Zhang, Honggang ;

Wang, Xiaogang ;

Tang, Xiaoou .

30TH IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2017), 2017, :6450-6458

[30] Asphalt Pavement Pothole Detection and Segmentation Based on Wavelet Energy Field [J].

Wang, Penghui ;

Hu, Yongbiao ;

Dai, Yong ;

Tian, Mingrui .

MATHEMATICAL PROBLEMS IN ENGINEERING, 2017, 2017

← 1 2 3 4 →