Convolutional Neural Network with Attention Module for Identification of Tunnel Seepage. Issue 11 (November 2022)

• Record Type: Journal Article
• Authors: Chen, Qian; Xiong, Chuanguo; Lv, Weishan; Shen, Ben; Zeng, Baoshan; Li, Jinming; Feng, Chenzefang; Hu, Zhou; Zhu, Fulong
• Is Part Of: Transportation research record. Volume 2676:Issue 11(2022)
• Page Start: 112
• Page End: 123
• Publication Date: 2022-11
• Subjects: seepage detection -- deep learning -- generative adversarial nets -- attention module; Transportation -- Periodicals; Roads; Transport -- Périodiques; Routes -- Périodiques; Routes -- Conception et construction -- Périodiques; Roads; Transportation; 388.05
• DOI: 10.1177/03611981221091774
• Languages: English
• ISSNs: 0361-1981

Abstract

As tunnel construction proceeds ever more rapidly, the efficiency of seepage detection by engineers with expert knowledge is facing unprecedented challenges. Moreover, it suffers from strong subjectivity. In recent years, deep learning, as an algorithm of machine learning, has achieved state-of-the-art performance in pattern recognition. In this paper, we address such a problem by building convolutional neural networks that operate on conventional graphics processing units. Within the project, the data is obtained by an infrared thermal imager since there exist different characteristics of temperature between the area of seepage and non-seepage. Considering the difficulty of collecting many images, generative adversarial nets and other data augmentation skills are applicable to enlarge data sets. We design several novel architectures where the attention mechanism is plugged into various traditional models, considered as VGG16 network with Attention Module and RestNet34 with Attention Module, and the overall identification accuracy achieved is more than 97%. The codes of this project can be found athttps://github.com/Scotter-Qian/cnn .