|Abstract or Summary
- Urban green space is associated with multiple physical and mental health outcomes.
Several benefits of green space, such as stress reduction and attention restoration, are
dependent on visual perception of green space exposures. However, traditional green
space exposure measures do not capture street-level exposures. In this project, we apply
deep learning models to measure green space in street view imagery. We train Faster
R-CNN model on PasadenaUrbanTree dataset and equip it with the ability to detect
trees, which is then used to count the number of trees in the test images. We also employ
a PSPNet model that pretrained on ade20k dataset to do semantic segmentation on
street view images and compute the portion of green space. Combining the outcomes
from object detection and semantic segmentation, the green space in street view
imagery can be measured quantitatively.