Yuequan Bao

TL;DR: Inspired by the real-world manual inspection process, a computer vision and deep learning–based data anomaly detection method is proposed that shows that the multi-pattern anomalies of the data can be automatically detected with high accuracy.

TL;DR: A brief review of the state of the art of data science and engineering in SHM as investigated by these authors covers the compressive sampling-based data-acquisition algorithm, the anomaly data diagnosis approach using a deep learning algorithm, crack identification approaches using computer vision techniques, and condition assessment approaches for bridges using machine learning algorithms.

TL;DR: A novel data anomaly detection method based on a convolutional neural network (CNN) that imitates human vision and decision making is proposed, which could detect the multipattern anomalies of SHM data efficiently with high accuracy.

TL;DR: Results show that the trained modified fusion convolutional neural network can automatically detect the cracks, handwriting, and background from the raw images and the recognition errors of the fusion convolved neural network in both the training and validation processes are smaller than those of the regular convolutionAL neural network.

TL;DR: The potential of CS for data compression of vibration data is investigated using simulation of the CS sensor algorithm and the values of compression ratios achieved are not high, because the vibration data used in SHM of civil structures are not naturally sparse in the chosen bases.