NUTRIENT DEFICIENCY DIAGNOSIS IN WHOLE HYDROPONIC LETTUCE BASED ON RANDOM FOREST

The phenotypic information of lettuce leaves can well reflect its health. In order to diagnose the nutrient deficiency types hydroponic accurately, non-destructively and quickly in mid-growth stage, a method for diagnosis whole based on random forest algorithm (RF) was proposed. images under four different conditions, K-deficiency, Ca-deficiency, N-deficiency Normal, were collected segmented by Extra-green algorithm. Then, features color, texture shape extracted. A RF classification model constructed compared with support vector machine (SVM) back propagation neural network (BP). had best effect among three methods. overall accuracy 86.32%, Kappa coefficient 0.82, it provide basis prevention remedies scientific management solutions.