Quality Classification of Dragon Fruits Based on External Performance Using a Convolutional Neural Network

Currently, most agricultural products in developing countries are exported to many around the world. Therefore, classification of these according different standards is necessary. In Vietnam, dragon fruit considered as with highest export rate. carried manually, lead low-quality high labor costs. this study describes an automatic classifying system using non-destructive measurements, based on a convolutional neural network (CNN). This uses combination model machine learning and image processing identify external features fruits; fruits then classified evaluated by groups. The recognized system, which extracts objects combined signal obtained from loadcell calculate determine each group. training data collected dataset images more than 1287 fruits, train model. speed accuracy two important factors. results show that achieves efficiency. effective existing types. Vietnamese factories, increases sorting capacity packing facilities six times higher manual method, 96%.