Implementasi Convolutional Neural Network Untuk Klasifikasi Daging Menggunakan Fitur Ekstraksi Tekstur dan Arsitektur AlexNet

The demand for meat began to increase rapidly, causing drastic price changes and the existence of scammers inflate get big profits by mixing beef pork. Few consumers are aware meat, distinguish between pork can be seen in terms color texture, but this theory still has weaknesses. This research uses Deep Learning method, namely Convolutional Neural Network with Local Binary Pattern texture extraction feature AlexNet architecture classification. conducted stated that accuracy image classification measured using various parameters optimizers. highest results obtained from study were 68.6% accuracy, 62% precision, 57.6% recall, 59% f1-score Stochastic Gradient Descent (SGD) optimizer, 0.01 learning rate, 32 batch size, 0.9 momentum. Compared original dataset, LBP dataset type is below 84.1% 78.6% 79% RMSprop 0 .0001 sizes, momentum So it concluded setting existing parameter values value.