Objective and non-destructive growth measurement of cabbage [Brassica oleracea capitata] plug seedlings population using machine vision

The aim of this study is to develop an objective and non-destructive method of growth measurement of cabbage plug seedlings using a machine vision. A method of discriminant analysis was used to select the threshold value of the chromaticity g component for the objective extraction of images. The soil coverage and standard deviation of lightness in the extracted images were calculated and inputted to a neural network model for the non-destructive measurement of the leaf area. The predicted leaf area of test plug seedlings populations based on the neural network model were fitted well with the actual values. The coefficient of determination R2 was 0.94. The leaf area of the cabbage plug seedlings population could be measured objectively and non-destructively using the machine vision.