Development of a digital image analysis method for real-time estimation of chlorophyll content in micropropagated potato plants

The present work describes a digital image analysis method based on leaf color analysis to estimate chlorophyll content of leaves of micropropagated potato plantlets. For estimation of chlorophyll content, a simple leaf digital analysis procedure using a simple digital still camera was applied in parallel to a SPAD chlorophyll content meter. RGB features were extracted from the image and correlated with the SPAD values. None of the mean brightness parameters (RGB) were correlated with the actual chlorophyll content following simple correlation studies. However, a correlation between the chromaticity co-ordinates ‘r’, ‘b’ and chlorophyll content was observed, while co-ordinate ‘g’ was not significantly correlated with chlorophyll content. Linear regression and artificial neural networks (ANN) were applied for correlating the mean brightness (RGB) and mean brightness ratio (rgb) features to chlorophyll content of plantlet leaves determined through a SPAD meter. The chlorophyll content as determined by the SPAD meter was significantly correlated (RMSE = 3.97 and 3.59, respectively, for linear and ANN models) to the rgb values of leaf image analysis. Both the models indicate successful prediction of chlorophyll content of leaves of micropropagated plants with high correlation. The developed RGB-based digital image analysis has the advantage over conventional subjective methods for being objective, fast, non-invasive, and inexpensive. The system could be utilized for real-time estimation of chlorophyll content and subsequent analysis of photosynthetic and hyperhydric status of the micropropagated plants for better ex vitro survival.