Using AI to Extract Biophilic Design Elements and Predict Health Benefits and Tradition Environmental Qi
This item is part of: Boarin, P., Haarhoff, E., Manfredini, M., Mohammadzadeh, M., Premier, A., (2021). Rethinking Sustainable Pacific Rim Territories. Proceedings of the 2020 APRU Sustainable Cities and Landscapes Hub PhD Symposium, Future Cities Research Hub, School of Architecture and Planning of the University of Auckland. ISBN: 978-0-473-53616-9
People release stress in urban environments by experiencing green areas, such as parks, grasslands, and areas with trees and hedges. For over 30 years, increasing studies have depicted the psychological and physiological health benefits of experiencing nature. However, recently, people have been staying in concrete environments without green spaces in their daily lives, especially during the COVID-19 pandemic, not only causing social isolation but also contributing to health problems. Biophilic attributes in built environments might improve people’s connection to nature and provide health benefits and influence landscape design applications. To confirm this, the present study took photos in urban green spaces and imported them into Google Vision AI to label their biophilic attributes and to predict the tradition environmental Qi in the space. The study found that natural labels, such as “tree, plant, grass, and park” significantly influenced people’s preference for a space, its tradition environmental qi, and people’s experiences of recovery and reflection. However, urban labels, such as “building, architecture, city, and house” were significantly negatively related to the same psychological outcomes. Using AI to define biophilic labels could optimize the psychological benefits of designed spaces and provide a new view for related landscape design work.