Artificial Neural Network based Model for Fruit Grade Empirical Thresholding and Feature Extraction based Back Propagation

Abstract

This study details a novel attribute retrieval method for use in pre-processing images, and then applies it to the development of an "artificial neural network" system based on back propagation for identifying fruits in photographs. The “Scale Conjugate Gradient” (SCG) technique is used For back propagation. In this paper, there are three stages to the process. First, MATLAB was used to process a variety of external image-based apple properties. Since merely colour is insufficient to judge the quality, size and weight characteristics were also taken into consideration. Second, features extraction was carried out during picture pre-processing to simplify the method by concentrating only on important features. The Support Vector Machine (SVM) algorithm is a favourite for creating classification models that are relatively small in weight. The classification in this work is done using the MATLAB-ANN (Artificial Neural Network) toolkit. A single hidden layer BP-ANN (Back propagation- artificial neural network) was employed with sigmoid activation functions,. The outcome was determined by the appropriate output variables, which is the apple's quality class, which was determined to be Class A, Class B, Class C, and Class D, respectively. The modeling result indicates the tremendous match between the data used in training and assumed output values. It also has shorter calculation time due to the SCG algorithm. It is also possible for apple producers and distributors to classify their fruit using this model and reduce the cost by avoiding manual classification.