Read the full-text of this article for free

Title: The 3D digital revitalisation of industrial heritage in energy sector for the third front construction in Southwest China

Authors: Yanlong Liu; Peiyun Cheng; Jie Li; Seokchool Kim

Addresses: School of Art, Chongqing Technology and Business University, Chongqing, 400067, China ' School of Art, Chongqing Institute of Foreign Studies, Chongqing, 401331, China ' School of Economics, Sichuan University of Science and Engineering, Zigong, 643000, China ' Tourism Management Department, PaiChai University, Daejeon, 340934, Korea

Abstract: This study proposes a deep learning (DL)-based multimodal three-dimensional (3D) digital revitalisation and regeneration method: DL-based heritage activation and reconstruction (DL-HAR). It aims to address the complexity and conservation needs of the Industrial heritage in energy sector from the Third Front Construction in Southwest China. First, the method constructs a 3D dataset based on laser scanning and drone imagery while implementing a dual-path DL model. Second, this method combines geometric feature extraction (PointNet++) with texture semantic understanding (ResNet-50) to align and dynamically repair cross-modal data. These results indicate that the DL-HAR method outperforms traditional methods and mainstream baseline models. Therefore, this study provides a theoretical basis and technical framework for the digital protection and revitalisation of Industrial heritage in energy sector, promoting a shift from static to dynamic cultural heritage conservation.

Keywords: industrial heritage in energy sector; DL-HAR; DL-based heritage activation and reconstruction; deep learning model; feature extraction; texture semantics.

DOI: 10.1504/IJISD.2026.155827

International Journal of Innovation and Sustainable Development, 2026 Vol.20 No.8, pp.1 - 24

Received: 11 Dec 2025
Accepted: 16 Feb 2026
Published online: 24 Aug 2026 *