Journal
COMPUTER-AIDED CIVIL AND INFRASTRUCTURE ENGINEERING
Volume 36, Issue 12, Pages 1568-1584Publisher
WILEY
DOI: 10.1111/mice.12753
Keywords
-
Categories
Funding
- Guangdong-HongKong-Macau Joint Laboratory Program [2020B1212030009]
- Science andTechnology Development Fund of the Macau SAR [SKL-IOTSC-20212023]
- Systematic Project ofGuangxiKey Laboratory ofDisaster Prevention and Engineering Safety [2020ZDK0011]
Ask authors/readers for more resources
The proposed FF-BLS crack detection system utilizes fusion features and incremental learning to achieve efficient training, resulting in similar recognition accuracy as well-known deep convolutional neural networks but with more than 20 times faster training speed.
Deep learning has been widely applied to vision-based structural damage detection, but its computational demand is high. To avoid this computational burden, a novel crack detection system, namely, fusion features-based broad learning system (FF-BLS), is proposed for efficient training without GPU acceleration. In FF-BLS, a convolution module with fixed weights is used to extract the fusion features of images. Feature nodes and enhancement nodes randomly generated by fusion features are used to estimate the output of the network. Meanwhile, the proposed FF-BLS is a dynamical system, which achieves incremental learning by adding nodes. Thus, the trained FF-BLS model can be updated efficiently with additional data, and this substantially reduces the training cost. Finally, FF-BLS was applied to crack detection. Compared with some well-known deep convolutional neural networks (VGG16, ResNet50, InceptionV3, Xception, and EfficientNet), the FF-BLS achieved a similar level of recognition accuracy, but the training speed was increased by more than 20 times.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available