@article{wang_zargar_yuan_2021, title={Augmented reality for enhanced visual inspection through knowledge-based deep learning}, volume={20}, ISSN={["1741-3168"]}, DOI={10.1177/1475921720976986}, abstractNote={A two-stage knowledge-based deep learning algorithm is presented for enabling automated damage detection in real-time using the augmented reality smart glasses. The first stage of the algorithm entails the identification of damage prone zones within the region of interest. This requires domain knowledge about the damage as well as the structure being inspected. In the second stage, automated damage detection is performed independently within each of the identified zones starting with the one that is the most damage prone. For real-time visual inspection enhancement using the augmented reality smart glasses, this two-stage approach not only ensures computational feasibility and efficiency but also significantly improves the probability of detection when dealing with structures with complex geometric features. A pilot study is conducted using hands-free Epson BT-300 smart glasses during which two distinct tasks are performed: First, using a single deep learning model deployed on the augmented reality smart glasses, automatic detection and classification of corrosion/fatigue, which is the most common cause of failure in high-strength materials, is performed. Then, in order to highlight the efficacy of the proposed two-stage approach, the more challenging task of defect detection in a multi-joint bolted region is addressed. The pilot study is conducted without any artificial control of external conditions like acquisition angles, lighting, and so on. While automating the visual inspection process is not a new concept for large-scale structures, in most cases, assessment of the collected data is performed offline. The algorithms/techniques used therein cannot be implemented directly on computationally limited devices such as the hands-free augmented reality glasses which could then be used by inspectors in the field for real-time assistance. The proposed approach serves to overcome this bottleneck.}, number={1}, journal={STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL}, author={Wang, Shaohan and Zargar, Sakib Ashraf and Yuan, Fuh-Gwo}, year={2021}, month={Jan}, pages={426–442} }
 @misc{zargar_yuan_2021, title={Impact diagnosis in stiffened structural panels using a deep learning approach}, volume={20}, ISSN={["1741-3168"]}, DOI={10.1177/1475921720925044}, abstractNote={Low-velocity impact on a structure emanates an elastic wave that propagates through the structure carrying a wealth of information about the impact event. This propagating wave can be visualized through a series of images (time-frames in the context of computer-vision) in the time–space domain collectively referred to as the wavefield. An approach for the autonomous analysis of these wavefields is presented in this article for the purpose of impact diagnosis, that is, identifying the impact location and reconstructing the impact force time-history. The high spatio-temporal dimensionality of the wavefield mandates the use of deep neural networks for analysis; however, unlike the traditional object detection problem in computer-vision, the nature of the impact diagnosis problem requires the capturing of context from the wavefield evolution. This necessitates learning across multiple time-frames of the wavefield simultaneously rather than focusing independently on each frame. While scanning simultaneously across multiple time-frames provides indispensable information about the wave propagation phenomenon in terms of its interactions with geometric features, boundaries, and so on, it mandates the use of deep learning models that can analyze this complex phenomenon in both spatial and temporal domains. A unified CNN-RNN network architecture is employed in this article to address this issue of spatio-temporal information extraction. The proposed approach is verified using simulated wavefields obtained from the finite element analysis of a five-bay stiffened aluminum panel. In order to demonstrate the generalization capabilities of the model, simulated wavefields corresponding to highly idealized impact scenarios are used for training, whereas for testing, the ones corresponding to more realistic impacts are used. It is shown that by incorporating the physics-based concept of time-reversal in the recurrent part of the network, better network performance can be achieved. The potential extension of the proposed methodology to an end-to-end vision-based impact monitoring system is also discussed at the end.}, number={2}, journal={STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL}, author={Zargar, Sakib Ashraf and Yuan, Fuh-Gwo}, year={2021}, month={Mar}, pages={681–691} }
 @article{yuan_zargar_chen_wang_2020, title={Machine Learning for Structural Health Monitoring: Challenges and Opportunities}, volume={11379}, ISSN={["1996-756X"]}, DOI={10.1117/12.2561610}, abstractNote={A physics-based approach to structural health monitoring (SHM) has practical shortcomings which restrict its suitability to simple structures under well controlled environments. With the advances in information and sensing technology (sensors and sensor networks), it has become feasible to monitor large/diverse number of parameters in complex real-world structures either continuously or intermittently by employing large in-situ (wireless) sensor networks. The availability of this historical data has engendered a lot of interest in a data-driven approach as a natural and more viable option for realizing the goal of SHM in such structures. However, the lack of sensor data corresponding to different damage scenarios continues to remain a challenge. Most of the supervised machine-learning/deep-learning techniques, when trained using this inherently limited data, lack robustness and generalizability. Physics-informed learning, which involves the integration of domain knowledge into the learning process, is presented here as a potential remedy to this challenge. As a step towards the goal of automated damage detection (mathematically an inverse problem), preliminary results are presented from dynamic modelling of beam structures using physics-informed artificial neural networks. Forward and inverse problems involving partial differential equations are solved and comparisons reveal a clear superiority of physics-informed approach over one that is purely datadriven vis-à-vis overfitting/generalization. Other ways of incorporating domain knowledge into the machine learning pipeline are then presented through case-studies on various aspects of NDI/SHM (visual inspection, impact diagnosis). Lastly, as the final attribute of an optimal SHM approach, a sensing paradigm for non-contact full-field measurements for damage diagnosis is presented.}, journal={SENSORS AND SMART STRUCTURES TECHNOLOGIES FOR CIVIL, MECHANICAL, AND AEROSPACE SYSTEMS 2020}, author={Yuan, Fuh-Gwo and Zargar, Sakib Ashraf and Chen, Qiuyi and Wang, Shaohan}, year={2020} }