Klasifikasi Citra Daging Sapi dan Daging Babi Menggunakan CNN Arsitektur EfficientNet-B6 dan Augmentasi Data

In daily life, beef often serves as a staple food for humans. However, the high and expensive price of has prompted traders to adulterate it with pork sake profit. Such adulteration serious implications in Islamic religion, where not all types meat are considered halal (permissible consumption), such pork. As result, consumers remain unaware that they purchase been adulterated At glance, both exhibit similar appearance texture, making them difficult differentiate. This research aims classify using deep learning model Convolutional Neural Network (CNN) method, combined data augmentation. The used is EfficientNet-B6 variations testing scenario. include ratio training data, rates, optimizer EfficientNet-B6. Data augmentation performed techniques random rotation, shifting, image scaling, vertical horizontal flipping, nearest pixel filling. Evaluation results confusion matrix show achieves highest accuracy classes beef, pork, samples at 92.00%, while without an 91.67%. from this experiment, best scenario avoid misclassifying can be obtained. involves augmentation, 90:10 split, SGD optimizer, rate 0.01, which precision class 96.05%. findings demonstrate use on images improve model's performance, 0.01 exhibits performance classifying images.