Название: The Rise of Smart Cities : Advanced Structural Sensing and Monitoring Systems
Автор: Amir H. Alavi, Maria Q. Feng, Pengcheng Jiao
Издательство: Butterworth-Heinemann/Elsevier
Год: 2022
Страниц: 698
Язык: английский
Формат: pdf (true)
Размер: 105.2 MB

The Rise of Smart Cities: Advanced Structural Sensing and Monitoring Systems provides engineers and researchers with a guide to the latest breakthroughs in the deployment of smart sensing and monitoring technologies. The book introduces readers to the latest innovations in the area of smart infrastructure-enabling technologies and how they can be integrated into the planning and design of smart cities.

With this book in hand, readers will find a valuable reference in terms of civil infrastructure health monitoring, advanced sensor network architectures, smart sensing materials, multifunctional material and structures, crowdsourced/social sensing, remote sensing and aerial sensing, and advanced computation in sensor networks.

Data analytics refers to the use of the appropriate statistical analysis methods to analyze (i.e., summarize, understand, and digest) a large amount of sensing and monitoring data. Data analytics is particuarly the process of studying and generalizing the monitoring data to obtain the health status of the civil infrastructures. It can be divided into data preprocessing, data fusion, pattern recognition, data processing, and data visualization steps.

Advancement in processing capability and availability of large datasets have led to the growing popularity of convolutional neural networks (CNN) in recent times. In keeping with the widespread use of CNN in other walks of life, many researchers in the SHM community explored its potential to be used as a tool for visual data analysis assisting in the smart autonomous inspection. Crack detection on the concrete surface has been one of the major use cases of CNN-based classifiers. Yokoyama and Matsumoto developed a CNN-based framework to distinguish concrete crack patches from a noncrack background in inspection images. Xu et al. incorporated atrous convolution, atrous spatial pyramid pooling, and depth-wise separable convolution to propose a CNN-based bridge crack identification approach.