基于人工智能技术的植被色彩感知与焦虑关联分析

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.