A review on fruit quality assessment through machine learning algorithms

The primary focus of this study is to examine the current advancements in fruit classification and disease detection. Fruits are a vital component of the human diet. A machine learning-based application can be developed to reduce human error in detecting and classifying fruit diseases. The commonly used algorithm is Convolution Neural Network (CNN) in feature extraction and detection. The variants of the CNN such as RestNet, AlexNet, MobileNet, and GoogleNet give better performance. The features such as texture, colour, size, and shape are analysed. The ancestor study discusses that inputted data is pre-processed so that the dataset will fit into the model. The data augmentation process will help to expand the dataset. The results obtained from applying the machine learning model to assess machine learning in the fruit quality assessment process will help improve precision and scalability.