Abstract:
Greenery visibility in urban environments provides numerous psychological and ecological
benefits. However, existing indices, such as the Green View Index and NDVI, fail to adequately
capture greenery as perceived from a pedestrian’s eye level. This study introduces a novel three
dimensional (3D) method for evaluating greenery visibility using stereo imagery and depth data
derived from eye-level photographs. The study was conducted in Ageo City, Saitama Prefecture,
Japan. Plant placement in frontage greening was evaluated using actual photographs. Frontage
greening in detached urban houses was captured from a stereo camera and 713 photographs
were analysed. A custom-developed program was used to extract depth values of green pixels
(Green Depth, GD). The GD value indicates the distance from the viewpoint to the green element
along the Z-axis direction. Parallel lines (one through the person, one through the tree) lie along
the X-axis, with the Z-axis perpendicular to the parallel reference lines aligned with the X-axis.
Using depth value, depth-based indicators were calculated and multiple regression analysis was
performed with façade image-based indicators that measure the degree of mixture of greenery
and artificial objects, such as greenery-to-greenery proximity, greenery-to-artificial-object
proximity and agglomeration. The multiple correlation coefficients were found for standard
deviation (0.30), kurtosis (0.35), Simpson diversity index (0.32), proportionality (0.38) and for
GD value average (0.48, skewness (0.51) and degree of openness (0.59). Finally, this novel
method was proposed to quantify greenery in three dimensions. As future work, using aerial
images, 2D greenery visibility indexes will be calculated using satellite images for the same
detached houses and will be compared with these 3D indicators. Integrating 2D and 3D indexes
will result in a methodology to evaluate and understand urban greenery visibility as people really
experience it. Furthermore, aerial photographs will be used to evaluate land plot sizes of the
detached houses and the positioning of the building, car parks, plants and entrances. This
approach provides urban planners and landscape designers with a more human-centered tool for
assessing and improving the visibility and distribution of greenery in urban neighborhoods.
