Acta Geographica Sinica ›› 2023, Vol. 78 ›› Issue (4): 1044-1056.doi: 10.11821/dlxb202304017

• Behavioral Geography • Previous Articles     Next Articles

The correlation between plant color perception and anxiety based on artificial intelligence technology

WU Jiayu1(), WANG Shiyi1, LI Hong2,3,4, TA Na2,3,4()   

  1. 1. Institute of Landscape Architecture, Zhejiang University, Hangzhou 310058, China
    2. Key Laboratory of Geographic Information Science, Ministry of Education, East China Normal University, Shanghai 200241, China
    3. School of Geographic Sciences, East China Normal University, Shanghai 200241, China
    4. Key Laboratory of Spatial-temporal Big Data Analysis and Application of Natural Resources in Megacities, Ministry of Natural Resources, Shanghai 200241, China
  • Received:2022-05-09 Revised:2022-12-28 Online:2023-04-25 Published:2023-04-13
  • Contact: TA Na;
  • Supported by:
    National Natural Science Foundation of China(32271935);National Natural Science Foundation of China(41971200);National Natural Science Foundation of China(51908488);Zhejiang Provincial Natural Science Foundation(LY22E080013);Young Elite Scientists Sponsorship Program by CAST(2021QNRC001)


The alleviating effect of urban green space on anxiety has attracted the attention of scholars from multiple disciplines such as urban geography, urban planning and landscape ecology. However, whether plant color, as an important feature of green space, plays a key role in alleviating anxiety has not been examined thus far due to the limitation of the accuracy of data collection technology. Accordingly, this research attempts to innovatively construct a system to describe the colors of plant so as to explore the relationship between plant color and residents' anxiety under different geographical backgrounds. Through the combination of public citizen science and artificial intelligence, we constructed a high-precision urban forestry database and subsequently evaluated plant color in two levels from the perspective of visual perception, and compared the difference of plant colors in residential areas and workplaces. Ordinal logistic regression models were established to explore the relationship between plant color and anxiety. Research shows that plant color diversity can effectively relieve anxiety, but it was influenced by modifiable areal unit problem (MAUP) and uncertain geographic context problem (UGCoP). Specifically, improving the base color level of the living environment and increasing the diversity of plant colors in the workplace can help alleviate anxiety, and the correlation between plant color and anxiety is mainly observed in small-scale buffer zones. The above results confirm the correlation between plant color and anxiety, and can provide specific optimization suggestions for urban green space design.

Key words: anxiety, artificial intelligence, citizen science, plant color, geographical context