Possibilities of using image analysis to estimate the nitrogen nutrition status of apple trees

Precise timing and rate of nitrogen (N) fertilizer application play a major role for crop yield and quality. Prediction of N requirements is necessary for efficient utilization of nitrogen fertilizers. A user-friendly method for estimating foliar nitrogen concentration, based on an analysis of leaf blade colour, was developed in the study. Apple tree (Malus domestica Borkh.) was used as a model species for the validation of the proposed method. Digital recording of adaxial leaf surface was performed using a flatbed scanner. The acquired images were analysed to obtain average values of basic colours: red, green and blue (RGB). Analyses of the relationships between N concentration in the leaves, leaf greenness indices (determined with SPAD-502 and CCM-200 meters) and RGB colour components were performed. The best fitted models were: a multiple regression model based on RGB primary colour values and a model that included the values of the red and blue colours as well as a parameter describing the age of the leaf. The performed tests showed that the analysis of scanned images provided a more precise determination of leaf nitrogen content in apple trees compared to the measurements conducted with leaf greenness meters.