Color analysis was found to be a low-cost and noninvasive technique to distinguish plant and soil, crop and weed, living and dead plant, diseased and healthy leaves, and human and plant in field and automatic seedling transplantation systems. However, color analysis was rarely utilized to differentiate various types of shade by grass cover. The main objective of this study was to quantify and differentiate mix grass cover under shade and light in a green infrastructure. The spatial heterogeneity of mix grass species in the selected site was categorized into three different portions, i.e., (1) mix grass cover under tree shade (MUT); (2) mix grass cover under self-shade (MUS); and (3) mix grass cover without shade (MWS). Field monitoring was conducted for about six months to validate the color analysis technique. A java-based image analysis tool, ImageJ, was used for color analysis. Lab color space was adopted for the present study, which identifies the spatial heterogeneity of mix grass cover based on lightness variation of green color. L describes lightness, and a and b represent the color opponents green–red and blue–yellow. ImageJ categorizes L, a, and b using 256 segments on a scale varying between 0 and 255. The range of L values for MUT, MUS, and MWS were found to lie within 20–91, 92–162, and 162–233, respectively. Ranges of a and b values were found to be 0–134 and 154–255, respectively, for the three categories of mix grass. Proportions of MUT, MUS, and MWS were found to vary within 0.2–3.0 %, 0–53 %, and 12–49 %, respectively. Variation in MUT and MWS proportions was found to be consistent with change in rainfall depth during the dry period. However, MUT and MWS were found to change with respect to shoot growth rate during the wet period. The MUS proportion variation trend was not found to be consistent with rainfall depth or shoot growth during the wet period.
CORE (COnnecting REpositories)
CORE (COnnecting REpositories)