Implementation of Growth Management System of Strawberries Using Statistical Regression Model and Image Processing Techniques

In this paper, we try to develop a growth management system that can maximize the harvest by examining frequently the growing status of fruit trees using statistical regression model and image processing technique. First, we segment the background, stem and strawberry body from a strawberry image. Second, we calculate the height and width of the strawberry by counting number of pixels from the grid given in the background. Third, a multiple regression model is used to derive a predictive function that can predict its weight from the given height and width of strawberries. Fourth, the accuracy of the prediction is verified by comparing the actual value with the predicted value of the weight of the strawberry using the verification data. Fifth, we conducted experiments using verification images to verify the performance of the constructed prediction model. From the experimental results, the value of the coefficient of determination for the constructed model is 0.98771, which is very high value. Finally, based on the algorithms derived so far, we have developed a platform that can actually be used to manage strawberry growth in farms.