Data fusion approaches for structural health monitoring and system identification: Past, present, and future

Abstract

During the past decades, significant efforts have been dedicated to develop reliable methods in structural health monitoring. The health assessment for the target structure of interest is achieved through the interpretation of collected data. At the beginning of the 21st century, the rapid advances in sensor technologies and data acquisition platforms have led to the new era of Big Data, where a huge amount of heterogeneous data are collected by a variety of sensors. The increasing accessibility and diversity of the data resources provide new opportunities for structural health monitoring, while the aggregation of information obtained from multiple sensors to make robust decisions remains a challenging problem. This article presents a comprehensive review of the recent data fusion applications in structural health monitoring. State-of-the-art theoretical concepts and applications of data fusion in structural health monitoring are presented. Challenges for data fusion in structural health monitoring are discussed, and a roadmap is provided for future research in this area.

A Survey on Multi-Scale Medical images Fusion Techniques: Brain Diseases

“…The reason for adapted Harvard datasets for this survey: (1) All images were co-registered accurately with different modalities (CT, MRI, PET, and SPECT) that used by researchers in the field of enhanced brain images fusion processing in the clinical /treatment analysis and evaluation performance perfectly. (2) It can show clearly the various types of diseases with different patients 'scenarios, for example ((Alzheimer's disease (AD) and brain attacks (stroke),) with highlight the specific abnormal part within brain like abnormal soft/hard tissues and abnormal cells.…”

“…The fusion technique, with a data fusion to organizes, connects, and combines multiple source and multi temporal data, gives the powerful tool for these data processing problems A complete robust and accurate image fusion scheme for application of brain diseases usually includes the following components corresponding to major steps of brain characterization process performed by a radiologist: (1) To quantify of lesions, diseases and/or regions of interest for the brain. (2) Lesions, diseases and/or region of interest characterization including specific feature extraction and decision making on the degree of malignancy and further course of action. (3) High accurate outcome brain medical images in term of (free noise images, visual quality of organs and tissues) are radiologist target.…”

“…Data fusion is the combination of different types of data to obtain more information from the merged data instead of considering each dataset separately [2]. It can be used to produce new raw data or more informative new data based on the original data.…”

Pruning deep convolutional neural networks for efficient edge computing in condition assessment of infrastructures

“…W. Park, Park, Kim, & Adeli, 2015;K. Park, Torbol, & Kim, 2018;Qarib & Adeli, 2014;Rafiei, Khushefati, Demirboga, & Adeli, 2017;Wu & Jahanshahi, 2018a), but current practice in structural health monitoring (SHM) still requires manual inspection made on the server are sent back to the edge device would be extremely inefficient, particularly because the transmission bandwidth is often limited. A preferable solution, which is referred to as edge computing (Shi et al, 2016), is to deploy edge devices that have the capability to analyze data, and make decisions about data acquisition without the support of a remote server.…”

“…According to the 2017 American Society of Civil Engineers (ASCE) infrastructure report (ASCE, ), the estimated cost for infrastructure rehabilitation, such as bridges, dams, and levees, requires billions of dollars. Early detection of deficiencies in structural components and surface defects can help in reducing the retrofit costs (Amezquita‐Sanchez & Adeli, ; Choi, Yeum, Dyke, & Jahanshahi, ; Kong & Li, ; Li, Park, & Adeli, ; Oh, Kim, Kim, Park, & Adeli, ; S. W. Park, Park, Kim, & Adeli, ; K. Park, Torbol, & Kim, ; Qarib & Adeli, ; Rafiei & Adeli, , ; Rafiei, Khushefati, Demirboga, & Adeli, ; Wu & Jahanshahi, ), but current practice in structural health monitoring (SHM) still requires manual inspection that is labor‐intensive and time‐consuming. The development of cost‐effective and autonomous SHM approaches is an urgent need in order to enhance the efficiency of inspection processes (Bertino & Jahanshahi, ).…”