Ensemble Convolutional Neural Network for Robust Batik Classification

EasyChair Preprint 5122

Abstract

Some researchers propose using the Convolutional Neural Network (CNN) method to classified batik images. It can extract features automatically without the need to define feature manually from the image. However, the weakness of CNN is that its accuracy is quite low, especially for small-sized datasets, when compared to machine learning methods that use hand-crafted feature extraction. In this research, an ensemble CNN method is proposed to improve the accuracy of the CNN method in classifying batik images. This method will train several CNN models at once, and then by voting and averaging techniques, the output label will be determined. Test results for two different datasets show this method can improve the accuracy of the CNN method and get an accuracy value of 100%. This method is also proven to be able to extract features faster than the MTCD+SVM method, which is included in the hand-crafted feature extraction category

Keyphrases: Batik, Batik Image Classification, Batik classification, Classification, Ensemble convolutional neural network, batik image, deep learning, ensemble cnn method, ensemble method

@booklet{EasyChair:5122,
  author    = {Yufis Azhar and Moch Chamdani Mustaqim and Agus Eko Minarno},
  title     = {Ensemble Convolutional Neural Network for Robust Batik Classification},
  howpublished = {EasyChair Preprint 5122},
  year      = {EasyChair, 2021}}