Classification and Prediction of Pineapple Quality using Deep Learning

Pineapple, a popular tropical fruit with widespread consumption, necessitates precise classification and quality prediction to ensure its marketability and profitability. Prior research has explored various methods, including thermal imaging and acoustic spectroscopy, to evaluate pineapple quality, but these methods often require costly equipment. In this research, deep learning techniques are used for pineapple classification and quality prediction. The primary objective is to develop a model that integrates VGG16, LSTM, and GAN techniques to classify pineapple maturity based on color, shape, and texture while utilizing YOLO for grading. The dataset used in this study comprises images of MD2 pineapples obtained from Malaysian farms. The developed prototype extracts pertinent features from images captured by established farm standards in Malaysia. Through an automated process, the model effectively classifies the quality of these pineapples.