OPTIMASI IMAGE CLASSIFICATION PADA JENIS SAMPAH DENGAN DATA AUGMENTATION DAN CONVOLUTIONAL NEURAL NETWORK

Garbage is useless goods/materials used normally or specifically in production, goods damaged during production materials which mainly come from households. Moreover, inorganic waste very difficult and takes a longer time to be decomposed by the soil. The lack of public knowledge about classification types how process it causes serious problem Indonesia. Therefore, this research creates type recognition program using Convolutional Neural Network (CNN) algorithm, can detect recognize objects an image. CNN technique inspired way mammals, humans, produce visual perception. included deep neural network because its high depth widely applied imagery. 2 Types classification, namely organic waste. implementation garbage image uses test models, Sequential on top VGG16 runs Google Collaboratory application, Keras. After carrying out Augmentation process, number data study was 1489 images training 182 testing resulting evaluation value with accuracy 90.97% loss 0.307 model, 97.99% 0.069 model. VGG16.