Coffee Bean Grading Based on Weight Estimation Using Densenet121 Model
Bipin Nair B J, Abrav Nanda K M, A. S. Shalwin, Likith Rao Mohethe G, V Raghavendra
Abstract
Coffee bean grading is a crucial process in the coffee industry, traditionally done manually by human experts, which is time-consuming, expensive, and prone to errors. To address these challenges, deep learning models have emerged as a powerful tool for automating coffee bean grading. This study uses the Densenet121 model to analyze a dataset of 363 images of manually graded coffee beans to recognize the different grades based on their features and patterns. The results show that the deep learning model achieves an accuracy of 81.89%, demonstrating its potential for revolutionizing the coffee industry by offering speed, accuracy, cost-effectiveness, scalability, and consistency in quality production. This study highlights the promising potential of deep learning models in image analysis and grading for various industries, with further research and development.