Title: Acoustic noise recognition of ancient buildings based on space time joint processing and deep learning
Authors: Ziqing Tang; Zhengguang Li
Addresses: Architecture College, Taiyuan University of Technology, Taiyuan 030024, China ' School of Civil Engineering and Architecture, Zhejiang University of Science and Technology, Hangzhou 310023, China
Abstract: Acoustic noise recognition of ancient buildings is crucial for the protection and study of ancient buildings, but the traditional methods have problems such as insufficient feature extraction and weak generalisation ability in complex scenes, which are ineffective for noise recognition. Therefore, this paper proposes an acoustic noise recognition method for ancient buildings based on space time joint processing and deep learning, which utilises space time joint processing to pre-process acoustic signals and extract effective features, and then classifies and recognises them through a deep learning model. Experiments show that the method shows excellent performance in terms of recognition accuracy and robustness, providing new ideas and effective means for the recognition of acoustic noise of ancient buildings, which helps to better protect and study ancient buildings.
Keywords: noise recognition; spatiotemporal joint processing; deep learning; acoustic signals; ancient architecture.
DOI: 10.1504/IJICT.2025.147140
International Journal of Information and Communication Technology, 2025 Vol.26 No.25, pp.70 - 86
Received: 16 May 2025
Accepted: 29 May 2025
Published online: 10 Jul 2025 *