Layanan penulisan ilmiah yang disediakan oleh Perpustakaan Universitas Gunadarma
IDENTIFIKASI GAMBAR DENGAN DATASET CIFAR-10 DENGAN MENGGUNAKAN ALGORITMA NEURAL NETWORK
ABSTRAKSI :
Digital image is an image that is represented as a numeric of a two-dimensional image. In its application, digital images have 3 types, to be able to predict images of each different type can use the Neural Network algorithm. This research uses google collaboration, tensorflow library, keras, numpy and CIFAR-10 dataset. In the results of image trials that have been carried out in this research has an image accuracy rate of 42.01% and a loss of 24.21%. This result is still relatively small for the level of accuracy, and for better results in the future studies, algorithm optimization can be used, and better data preparation is performed.