Klasifikasi Citra Multi-kelas Menggunakan Convolutional Neural Network

Authors

DOI:

https://doi.org/10.34818/eoe.v6i1.8501

Abstract

Abstrak ILSVRC (ImageNet Large Scale Visual Recognition Challenge) adalah suatu kompetisi tahunan yang melombakan klasiï¬kasi basis data gambar yang ada pada ImageNet yang memiliki 1000 kelas dan setiap tahunnya terdapat berbagai macam arsitekturjaringan convolutional neural network (CNN) yang menjadi state of the art. CNN digunakan karena metode ini bekerja dengan sangat baik dengan volume data yang besar. Dari 1000 kelas, akan diambil 100 kelas yang akan dijadikan dataset penelitian untuk mengetahui strategi pembelajaran seperti apa pada CNN yang dapat memiliki performa terbaik dengan berbagai skenario. Skenario terbaik didapatkan dengan melakukan training dengan ukuran gambar yang kecil, lalu melakukantraining kembalidenganmemperbesarukuran gambar. Skenarioinimendapatkanakurasi sebesar 75.82%, akurasi yang cukup tinggi untuk model yang dapat mengklasiï¬kasikan 100 kelas. Skenario ini juga memiliki performa yang paling baik dalam klasiï¬kasi keseluruhan berdasarkan ukuran evaluasi confusionmatrix. Katakunci: convolutionalneuralnetwork,multi-kelas,klasiï¬kasi,gambar Abstract ILSVRC (ImageNet Large Scale Visual Recognition Challenge) is a yearly competition that competes on classifyingImageNetimagedatabasethathas1000classesandeveryyear,anewarchitectureofconvolutional neural network (CNN) that becomes a state of the art emerge. CNN is used because this method works very well on a large volume of data. From 1000 classes available on ImageNet, 100 class will be used in thisresearchdatasettoknowwhichlearningstrategyforCNNperformedbestinthevaryingscenario. The best scenario is achieved by training the image in smaller resolution then training it again with a larger resolution. This scenario achieved an accuracy of 75.82%, it’s quite high for a model that can classify 100 class. It’salsoperformedbestaccordingtotheevaluationoftheconfusionmatrixgeneratedbyeachmodel. Keywords: convolutionalneuralnetwork,multi-class,classiï¬cation,image

Downloads

Download data is not yet available.

References

Imagenet Large Scale Visual Recognition Challenge (ILSVRC). Diakses 9 April 2018.

S. R. Amendolia, G. Cossu, M. Ganadu, B. Golosio, G. Masala, and G. M. Mura. A comparative study of k-nearest neighbour, support vector machine and multi-layer perceptron for thalassemia screening. Chemo metrics and Intelligent Laboratory Systems, 69(1-2):13–20, 2003.

I. Goodfellow, Y. Bengio, and A. Courville. Deep Learning. MIT Press, 2016. http://www. deeplearningbook.org.

I. Guyon, J. Weston, S. Barnhill, and V. Vapnik. Gene selection for cancer classification using support vector machines. Machine learning, 46(1-3):389–422, 2002.

K. He, X. Zhang, S. Ren, and J. Sun. Deep residual learning for image recognition. arXiv:1512.03385 [cs], Dec 2015. arXiv: 1512.03385.

B. Hu, Z. Lu, H. Li, and Q. Chen. Convolutional neural network architectures for matching natural language sentences. In Advances in neural information processing systems, pages 2042–2050, 2014.

Y. Hu, A.E.G.Huber, J. Anumula, and S. Liu. Overcoming the vanishing gradient problem in plain recurrent networks. CoRR, abs/1801.06105, 2018.

S. Ioffe and C. Szegedy. Batch normalization: Accelerating deep network training by reducing internal covariate shift. arXiv:1502.03167 [cs], Feb 2015. arXiv: 1502.03167.

S. Ji, W. Xu, M. Yang, and K. Yu. 3d convolutional neural networks for human action recognition. IEEE transactions on pattern analysis and machine intelligence, 35(1):221–231, 2013.

A. Krizhevsky, I. Sutskever, and G. E. Hinton. Imagenet classification with deep convolutional neural ne tworks. Communications of the ACM, 60(6):84–90, May 2017.

Y. Lecun, L. Bottou, Y. Bengio, and P. Haffner. Gradient-based learning applied to document recognition. Proceedings of the IEEE, 86(11):2278–2324, Nov. 1998.

L. Perez and J. Wang. The effectiveness of data augmentation in image classification using deep learning, 2017.

K. Simonyan and A. Zisserman. Very deep convolutional networks for large-scale image recognition. arXiv:1409.1556 [cs], Sep 2014. arXiv: 1409.1556.

L. N. Smith. Cyclical learning rates for training neural networks. CoRR, abs/1506.01186, 2015.

C. Szegedy, W. Liu, Y. Jia, P. Sermanet, S. Reed, D. Anguelov, D. Erhan, V. Vanhoucke, and A. Rabinovich. Going deeper with convolutions. arXiv:1409.4842 [cs], Sep 2014. arXiv: 1409.4842.

Downloads

Published

2019-04-01

Issue

Pages

2127-2136

Section

Program Studi S1 Informatika

How to Cite

Mahmud, K. H., Adiwijaya, A., & Faraby, S. A. (2019). Klasifikasi Citra Multi-kelas Menggunakan Convolutional Neural Network. EProceedings of Engineering, 6(1), 2127-2136. https://doi.org/10.34818/eoe.v6i1.8501