Deep Learning-Based Feature Silencing for Accurate Concrete Crack Detection

Sensors

HungManh La¹  Alireza Tavakkoli¹  UmmeHafsa Billah¹ 

[1] Department of Computer Science and Engineering, University of Nevada, Reno, NV 89557, USA;

关键词: convolutional neural network;    encoder-decoder architecture;    semantic segmentation;    feature silencing;    crack detection;   

DOI  :  10.3390/s20164403

来源: DOAJ

【 摘 要 】

An autonomous concrete crack inspection system is necessary for preventing hazardous incidents arising from deteriorated concrete surfaces. In this paper, we present a concrete crack detection framework to aid the process of automated inspection. The proposed approach employs a deep convolutional neural network architecture for crack segmentation, while addressing the effect of gradient vanishing problem. A feature silencing module is incorporated in the proposed framework, capable of eliminating non-discriminative feature maps from the network to improve performance. Experimental results support the benefit of incorporating feature silencing within a convolutional neural network architecture for improving the network’s robustness, sensitivity, and specificity. An added benefit of the proposed architecture is its ability to accommodate for the trade-off between specificity (positive class detection accuracy) and sensitivity (negative class detection accuracy) with respect to the target application. Furthermore, the proposed framework achieves a high precision rate and processing time than the state-of-the-art crack detection architectures.

【 授权许可】

Unknown