Classification of Papaya Fruit Maturity Level Using Image Processing

Abstract

PDF

Computer vision has seen great developments in the recent past, and they have facilitated broad applications across all aspects of life. One of the areas of application is classification of fresh produce, however, classifying vegetables and fruits has proven to be a challenging endeavour and requires careful construction. Fruit and vegetable classification is a challenging problem because it has similarities across classes and non-regular intraclass features. To create a papaya fruit classification system for evaluating quality, however the state-of-the-art has only been created for a restricted number of classes and datasets. Optimize algorithms for speed and efficiency, enabling real-time analysis in applications like surveillance or autonomous driving. Accurate high-quality, representative datasets and use data augmentation to improve robustness. To overcome the challenges of the CNN model proposed the model named YOLO v8 which provides the solution for different classification methods The goal of this study is to determine the status of papaya fruit, like matured, partially(semi) matured or unmatured. The model that was trained accomplished a precision recall accuracy as 98.2% and precision confidence achieved is 94% proving that this strategy is feasible. As per the results classification of mature, semimature and unmature results are 91%, 98 to 99%, 93 to 94 % respectively as improved compared with existing model.