When a potato tuber is exposed to light, its skin turns green because of the production of chlorophyll in the periderm. The extent of greening in potatoes is currently evaluated by visual inspection or by quantification of chlorophyll content. However, visual inspection is subjective and chemical quantification of chlorophyll content, although objective, requires time and labor. Moreover, because chemical analysis is a destructive procedure, it degrades potato tubers needed for propagation in breeding programs. Therefore, methods are needed to determine the extent of greening both quickly and nondestructively. The objective of this study was to develop a technique for estimating chlorophyll content from a photograph of the potato skin obtained with a smartphone camera. We developed a model to estimate chlorophyll content from the grayscale levels of the red, green, and blue components of the image. The coefficient of determination of the regression between the estimate and the chlorophyll content obtained by chemical analysis was 0.95. Estimation of the anthocyanin content, which also influences skin color, by the same method yielded a coefficient of determination of 0.84. Using an augmented reality technique, we developed a smartphone application that can display the contents of these two pigments in potato tubers and thus indicate the extent of greening.

View full abstract