Open AccessImage Classification of Pandawa Figures Using Convolutional Neural Network on Raspberry Pi 4
Author(s) -
Kartika Wisnudhanti,
Feri Candra
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1655/1/012103
Subject(s) - convolutional neural network , computer science , raspberry pi , artificial intelligence , convolution (computer science) , pattern recognition (psychology) , contextual image classification , image (mathematics) , layer (electronics) , test data , artificial neural network , architecture , network architecture , data mining , geography , world wide web , chemistry , computer security , organic chemistry , archaeology , programming language , internet of things
Pandawa is one of the stories in wayang show consisting of five figures: Yudhistira, Bima, Arjuna, Nakula and Sadewa. This research is using Convolutional Neural Network (CNN) and apply it on the Raspberry Pi 4 to classify the puppet figures. CNN is one of the methods that can be used for classification of image data that has more than two classes. The network architecture used can classify Pandawa figures, using 1000 dataset with a size of 100 × 100 consisting of 80% training data and 20% test data. The network architecture has 3 convolution layers and 3 hidden layers, as well as 1 layer output. The classification results on the training data have an accuracy rate of 97.88% while test data has an accuracy rate of 96.5%.