Prediction and Classification of Textural Properties on Fruits During Ripening Using Convolutional Neural Network

Abstract

Non-destructive quality detection and automatic grading are significant in fruit industry. The customary way separates fruits into four level ripening stages dependent on color. The assurance of the ripeness condition of fruits is a fundamental component in the agribusiness research field. In this paper a novel Convolutional Neural network is proposed for prediction and classification of textural properties on fruits during ripening. A PC vision system for ripening classification of fruits is acknowledged commonly dependent on a few cycles. Results and the exhibition of the proposed system are contrasted and different methods, for example, the ANN and SVM. Results uncover that the proposed system has the most noteworthy generally acknowledgment rate, which is 97.5%, among different methods.