Research on mural damage recognition method based on deep reinforcement learning

Jing Qiao; Xi Zhou; Min Li

doi:10.1117/12.3055444

12 December 2024 Research on mural damage recognition method based on deep reinforcement learning

Jing Qiao, Xi Zhou, Min Li

Proceedings Volume 13439, Fourth International Conference on Testing Technology and Automation Engineering (TTAE 2024); 134392F (2024) https://doi.org/10.1117/12.3055444
Event: Fourth International Conference on Testing Technology and Automation Engineering (TTAE 2024), 2024, Xiamen, China

Abstract

In this paper, the AE signal collected in the acoustic emission experiment of the mural collection is taken as the research object. The collected continuous mural state data are analyzed and processed, and the abnormal state information of mural cracks and falling off is extracted to obtain the sample set of disease characteristics. Wavelet transform was used to convert the collected AE signal of crack damage into a two-dimensional time-frequency image. Finally, deep reinforcement learning DQN network was used to identify the damage type. Compared with convolutional neural network, the test results showed that the damage recognition accuracy rate could reach 98%.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Jing Qiao, Xi Zhou, and Min Li "Research on mural damage recognition method based on deep reinforcement learning", Proc. SPIE 13439, Fourth International Conference on Testing Technology and Automation Engineering (TTAE 2024), 134392F (12 December 2024); https://doi.org/10.1117/12.3055444

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available