Urban parks have well-documented health benefits for urban residents. To increase the use of parks and enhance the physical activity level of city-dwellers, recent studies have explored the link between the amount of greenery in parks and the level of park usage. However, the results have been inconsistent, partly due to different measurements of park greenery. In this study, we developed a novel method to assess both the quantity and quality of park greenery from eye-level photographs taken along major paths in parks. The quantity of greenery in these photographs was objectively assessed with advanced machine learning techniques (PSPNet), and the quality of greenery was assessed by virtual audit. The associations between greenery and usage of 43 urban parks were further explored with regression models. The results showed that the quality of greenery has stronger association with total number of park visitors than the quantity. Both the quantity and quality of greenery had stronger associations with the number of elderly visitors (apparent aged 65 or above) than with the numbers of children or adults. Our results bring new insights into how park greenery can encourage park usage and contribute to healthy cities.

城市公园为城市居民带来了有据可查的健康益处。为了增加公园的利用并提高城市居民的身体活动水平，最近的研究探索了公园中的绿化量与公园使用水平之间的联系。但是，结果不一致，部分原因是公园绿化的测量不同。在这项研究中，我们开发了一种新颖的方法，可以从沿公园主要路径拍摄的与眼睛水平的照片上评估公园绿地的数量和质量。这些照片中的绿化量通过先进的机器学习技术（PSPNet）进行了客观评估，绿化的质量通过虚拟审核进行了评估。通过回归模型进一步探讨了绿色植物与43个城市公园的使用之间的关联。结果表明，绿化质量与公园游客总数的相关性强于数量。绿化的数量和质量与年长游客（65岁或以上的老年人）的数量之间的关联比与儿童或成人的数量之间的关联更强。我们的结果为公园绿化如何促进公园使用并为健康城市做出贡献提供了新见解。