Rachis browning is an important quality parameter of table grapes which can limit the development of novel storage technologies.
A web-based application was established to upload images and retrieve rachis-browning data.
Images acquired from photography or scanning were analyzed using digital image-processing algorithms to extract quantitative features of the rachis (Lichter et al. 2011). Briefly, the image color space was converted from RGB to hue saturation intensity (HSI). Then the rachis was segmented from the background by applying a threshold to the saturation component of S = 0.3. The extent of rachis browning was measured by segmenting the rachis pixels into two color classes: brown pixels were defined as pixels that had a hue value lower than a hue threshold (Hthr) of 50 while the rest of the pixels, with hue values above Hthr, were associated with fresh green rachis. The image-processing algorithms were coded in custom software that was written in C++, under the Microsoft Studio NET environment.