Perbandingan Kinerja Jaringan Saraf Tiruan Model Backpropagation dan General Regression Neural Network Untuk Mengidentifikasi Jenis Daging Sapi

Abstract

The research on image identification has been conducted to identify the type of beef. The research is aimed to compare the performance of artificial neural network of backpropagation and general regression neural network model in identifying the type of meat. Image management is processed by counting R, G and B value in every meat image, and normalization process is then carried out by obtaining R, G, and B index value which is then converted from RGB model to HSI model to obtain the value of hue, saturation and intensity. The resulting value of image processing will be used as input parameter of training and validation programs. The performance of GRNN model is more accurate than the backpropagation with accuracy ratio by 51%.