Objective In the context of healthy street, street plant space, as an integral component of street environment, has a significant influence on walking pleasure. Street plants or green spaces serve as the focus of the current researches on the relationship between walking pleasure and street plant area. At present, there is still a lack of systematic quantitative researches on and specific approaches for the evaluation of street plant space based on walking pleasure. Besides, there are few researches on the relationship between street plant space and walking pleasure. Therefore, it is necessary to discuss the characteristic evaluation systems for street plant space and walking pleasure, and to analyze the influence of street plant space on walking pleasure. Enriching the strategies for evaluation and design of street plant space based on walking pleasure can help promote the development and construction of healthy cities and healthy streets in China.

Methods Through the comparative analysis of nine sample streets in Wuhan, including living, commercial and traffic streets, this research builds evaluation systems for street plant space and walking pleasure in Wuhan with the correlation there between being fully considered. Specifically, the research first establishes the evaluation system for the morphological characteristics of street plant space through literature review and judgment matrix methods. Then, the research screens relevant characteristics of street plant space and walking pleasure by literature review and expert methods. Additionally, the research collects sample streetscape images through field research, and obtains sample street panorama photos using PTGui and Photoshop software. Moreover, the research obtains data about the morphological characteristics of street plant space by grid method and image segmentation method. After that, the research collects characteristic data about walking pleasure by means of experiment, questionnaire and interview, and explores the influence of street plant space on walking pleasure by such methods as correlation analysis, stepwise regression analysis and random forest. Finally, based on theoretical analysis, the research proposes the evaluation system and optimization design strategy for street plant space based on walking pleasure.

Results The results show that: 1) The evaluation system for morphological characteristics of street plant space can be divided into three dimensions: scale characteristics, shape characteristics and interface characteristics. Specifically, scale characteristic indicators include area (S) and height-width ratio (D/H); shape feature indicators include shape coefficient (K); interface characteristic indicators include side Ⅰ intervisibility (P_Ⅰ), side Ⅱ intervisibility (P_Ⅱ), top intervisibility (P_T) and visible green index (V_G). 2) The evaluation system for walking pleasure includes such indicators as satisfaction, relaxation, comfort, security and attraction, respectively accounting for 0.253 7, 0.237 9, 0.214 4, 0.162 9 and 0.131 1, which can be obtained by the judgment matrix method. Satisfaction has the greatest influence on walking pleasure. 3) The evaluation of walking pleasure of street plant space shows that sample 2 and sample 7 have the highest scores, both higher than 1; the walking pleasure score of sample 4 is the lowest, lower than 0. 4) The evaluation indicators of street plant space based on walking pleasure V_G, P_T, S, K, D/H and P_Ⅰ, whose influence weight value are respectively 0.288 5, 0.238 8, 0.184 1, 0.164 2, 0.099 5 and 0.024 9. V_G and P_T are the most important factors that should be considered in the evaluation of walking pleasure, which have the highest influence weight, followed by S and K, D/H and P_Ⅰ in succession.

Conclusion According to previous researches, there exists a nonlinear relationship between the influence of street plant space and walking pleasure. Therefore, the research adopts the random forest approach to determine the weight of assessment indicators of street plant space based on walking pleasure (V_G: 0.288 5, P_T: 0.238 8, S: 0.184 1, K: 0.164 2, D/H: 0.099 5, P_Ⅰ: 0.024 9). With the humanization concept as the leading concept, the research proposes an evaluation system for street plant space in megacities based on relevant research results and research interviews. In addition, the research formulates a clear standard to evaluate street plant space with a score range of 0-2. Specifically, a score higher than 1.5 indicates a high construction level; a score between 1.0 and 1.5 indicates an average construction level; a score lower than 1.0 indicates a poor construction level that needs to be further optimized. On this basis, the research proposes four optimization design strategies for street plant space. 1) Street space should be constructed or separated by plant elements, for which the management of tree spacing is of crucial importance. 2) The visible green index of street plant space should be kept within a reasonable range. 3) The complexity and richness of boundary shape of street plant space should be paid attention to. 4) The height-width ratio of street plant space should be kept between 0.25 and 1.00. Furthermore, the research puts forward relevant evaluation methods and optimization design strategies for street plant space based on walking pleasure, which can provide a reference for the construction of healthy streets.