Ripeness Classification of Lycopersicon Esculentum Employing Faster Region-Based Convolutional Neural Network with Inception Version 2

Accurate determination of tomato ripeness is essential for their marketability and consumer acceptance. In this study, the use of Convolutional Neural Networks for Object Detection particularly the Faster Region-Based Convolutional Inception v2 Architecture neural networks was explored for the detection and classification of tomato ripeness into three categories: Mature Green Tomato, Breaker Tomato, and Red Tomato. The researchers collected and preprocessed a dataset of tomato images, then the model was trained and evaluated using various detection and classification metrics. It demonstrated an accuracy of 78.57% for Mature Green, 88.37% for Breaker Tomato, and 100.00% for Red Tomato. In summary, the model performs well on test datasets with an accuracy percentage of 89.06%. The outcomes indicate how well the method works for precisely detecting and classifying tomato ripeness, with potential applications in commercial tomato production.