Quantifying the green view indicator for assessing urban greening quality: An analysis based on Internet-crawling street view data

The quality and aesthetic ecosystem services of urban green spaces can be assessed by green view (GV), which is an indicator to quantify the percentage of green area visually sensed by human eyes. There are several case studies on estimating GV at city-scale, however, the relationship between GV values and socio-economic & morphologic profiles of cities was rarely discussed. In this work, we analyzed the GV values and their potentially influential factors by adopting an internet data crawling approach, obtaining 36,654 panoramic Street View images from Baidu Map, in the urban zones at the Pearl River Delta Urban Agglomeration (PRDUA), China. We calculated the GV value of each panoramic image using MATLAB 2015a, extracting the value of the hue channel from the digital image and then calculating pixel ratio according the color spectrum. Results show that: (1) The overall GV in PRDUA is 11.3 +/- 7.5%, lower than the cases of the developed countries; (2) More green could be sensed on the streets at the old central districts and the central business district in the cities in PRDUA; (3) The GV value is positively affected by public revenue per unit area, but is uncorrelated to green space coverage ratio, at the district-level. It indicates that an improved design and increasing monetary investment of urban green space may obtain higher people-oriented green quantity in cities. These findings can be useful for drafting more appropriate urban green space planning policies regarding the green view.