Probability Voting-based Ensemble of Convolutional Neural nets Classifiers for Image Classification
Sarwo1, Yaya Heryadi2, Widodo Budiharto3, Edi Abdurachman4

1Sarwo*, Computer Science Department, BINUS Graduate Program  Doctor of Computer Science, Bina Nusantara University, Jakarta, Indonesia.
2Yaya Heryadi, Computer Science Department, BINUS Graduate Program Doctor of Computer Science, Bina Nusantara University, Jakarta, Indonesia
3Widodo Budiharto, Computer Science Study Program, School of Computer Science, Bina Nusantara University, Jakarta, Indonesia.
4Edi Abdurachman, Computer Science Department, BINUS Graduate Program – Doctor of Computer Science, Bina Nusantara University, Jakarta, Indonesia.

Manuscript received on 15 September 2022. | Revised Manuscript received on 15 September 2022. | Manuscript published on 30 September 2022. | PP: 60-68 | Volume-8 Issue-3 September 2019 | Retrieval Number: C3876098319/19©BEIESP | DOI: 10.35940/ijrte.C3876.098319
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: This study explores an ensemble technique for building a composite of pre-trained VGG16, VGG19, and Resnet56 classifiers using probability voting-based technique. The resulted composite classifiers were tested to solve image classification problems using a subset of Cifar10 dataset. The classifier performance was measured using accuracy metric. Some experimentation results show that the ensemble methods of pre-trained VGG19-Resnet56 and VGG16-VGG19-Resnet models outperform the accuracy of its individual model and other composite models made of these three models.
Keywords: Ensemble Classifiers, VGG16, VGG19, Resnet56, Probability Voting Technique, CIFAR-10.
Scope of the Article: