Toward sustainable and differentiated protection of cultural heritage illustrated by a multisensory analysis of Suzhou and Kyoto using deep learning

A novel approach using deep learning reveals significant differences in multisensory perceptions of cultural landscapes between Suzhou, China, and Kyoto, Japan. Unlike traditional studies focusing solely on visual elements, this research analyzes both images and text from online reviews to extract landscape features and sensory information across multiple senses. Findings indicate that different senses influence overall perception variably, providing new insights for integrating intangible heritage into sustainable protection strategies for culturally significant green spaces. This method offers a technological advancement in quantifying non-visual sensory elements, bridging the gap between tangible and intangible heritage preservation.

Zhang, J., Shi, Y., Zhao, L., Cai, C., & Furuya, K. (2025). Toward sustainable and differentiated protection of cultural heritage illustrated by a multisensory analysis of Suzhou and Kyoto using deep learning. Npj Heritage Science, 13(1). https://doi.org/10.1038/s40494-025-01855-z