<![CDATA[Beef is an essential food commodity with high economic value and a primary source of protein for society. The quality of beef affects consumer preferences, pricing, and market competitiveness. Quality assessment is generally conducted manually through visual inspection and smell, but this method is subjective and time-consuming and requires trained experts. This study aims to design and develop a beef quality classification system using a Convolutional Neural Network (CNN) model based on digital imagery. The dataset used consists of three beef quality categories: Grade 1 (fresh beef), Grade 2 (beef stored at room temperature for 7-14 hours), and Grade 3 (beef stored at room temperature for more than 14 hours). The dataset includes 180 images processed using cropping, resizing, and data augmentation techniques to enhance model variation and accuracy. The CNN architecture employs four convolutional layers with max pooling, followed by dropout and fully connected layers. The model was trained using the Adam optimizer, with a training-to-test data ratio of 80:20. Evaluation results demonstrated the model achieved an accuracy of 97.22%, with precision, recall, and f1-score values of 97.44%, 97.22%, and 97.22%, respectively. These findings suggest that the developed system has the potential to be used as an automatic tool for objective, fast, and accurate beef quality assessment.]]>
Copyrights © 2025
