Implementasi Data Augmentation Random Erasing dan GridMask pada CNN untuk Klasifikasi Batik

Chan Uswatun Khasanah, Angkin Kusuma Pertiwi, Farrel Witamajaya

Abstract


Penelitian ini bertujuan untuk mengetahui pengaruh data augmentation Random Erasing dan GridMask pada klasifikasi batik dengan 550 gambar yang terbagi menjadi 5 kelas, yaitu Ceplok, Kawung, Lereng, Nitik, dan Parang. Dataset terbagi menjadi data train, validation, dan test dengan perbandingan 70% : 20% : 10% sehingga masing-masing data train dan validation terdiri dari 500 gambar dan jumlah data test adalah 50 gambar. Kami mengimplementasikan data augmentation Random Erasing dan GridMask pada model pre-trained VGG16 dengan metode transfer learning dan fine-tuning (melakukan training pada block5 convolutional layer).

Berdasarkan training menggunakan model pre-trained VGG16 pada dataset batik dengan membandingkan metode fine-tuning dan transfer learning menunjukkan bahwa metode fine-tuning menghasilkan akurasi lebih tinggi daripada transfer learning. Training tanpa data augmentation dapat menghasilkan akurasi yang lebih tinggi daripada saat menggunakan data augmentation, namun masih mengalami overfitting. Overfitting tersebut dapat diperkecil dengan mengimplementasikan data augmentation Random Erasing dan GridMask.


Keywords


Klasifikasi batik; VGG16; overfitting; data augmentation

Full Text:

PDF (Indonesian)

References


R. L. Galvez, A. A. Bandala, E. P. Dadios, R. R. P. Vicerra, and J. M. Z. Maningo, “Object Detection Using Convolutional Neural Networks,” in IEEE Region 10 Annual International Conference, Proceedings/TENCON, 2019, vol. 2018-October. doi: 10.1109/TENCON.2018.8650517.

N. Jmour, S. Zayen, and A. Abdelkrim, “Convolutional neural networks for image classification,” 2018. doi: 10.1109/ASET.2018.8379889.

D. Liao, H. Lu, X. Xu, and Q. Gao, “Image Segmentation Based on Deep Learning Features,” 2019. doi: 10.1109/ICACI.2019.8778464.

A. J. Rozaqi, A. Sunyoto, and M. rudyanto Arief, “Deteksi Penyakit Pada Daun Kentang Menggunakan Pengolahan Citra dengan Metode Convolutional Neural Network,” Creative Information Technology Journal, vol. 8, no. 1, 2021, doi: 10.24076/citec.2021v8i1.263.

X. Liu, T. Kawanishi, X. Wu, and K. Kashino, “Scene text recognition with high performance CNN classifier and efficient word inference,” in ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, 2016, vol. 2016-May. doi: 10.1109/ICASSP.2016.7471891.

J. Tristanto, J. Hendryli, and D. Erny Herwindiati, “Classification of Batik Motifs Using Convolutional Neural Networks,” SSRN Electronic Journal, 2018, doi: 10.2139/ssrn.3258935.

T. Handhayani, J. Hendryli, and L. Hiryanto, “Comparison of shallow and deep learning models for classification of Lasem batik patterns,” in Proceedings - 2017 1st International Conference on Informatics and Computational Sciences, ICICoS 2017, 2017, vol. 2018-January. doi: 10.1109/ICICOS.2017.8276330.

I. M. A. Agastya and A. Setyanto, “Classification of Indonesian batik using deep learning techniques and data augmentation,” 2018. doi: 10.1109/ICITISEE.2018.8720990.

Y. Gultom, A. M. Arymurthy, and R. J. Masikome, “Batik Classification using Deep Convolutional Network Transfer Learning,” Jurnal Ilmu Komputer dan Informasi, vol. 11, no. 2, 2018, doi: 10.21609/jiki.v11i2.507.

A. Y. Wicaksono, N. Suciati, C. Fatichah, K. Uchimura, and G. Koutaki, “Modified Convolutional Neural Network Architecture for Batik Motif Image Classification,” IPTEK Journal of Science, vol. 2, no. 2, 2017, doi: 10.12962/j23378530.v2i2.a2846.

B. Ghojogh and M. Crowley, “The Theory Behind Overfitting, Cross Validation, Regularization, Bagging, and Boosting: Tutorial,” May 2019.

A. Krizhevsky, I. Sutskever, and G. E. Hinton, “ImageNet classification with deep convolutional neural networks,” Commun ACM, vol. 60, no. 6, 2017, doi: 10.1145/3065386.

K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” in Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2016, vol. 2016-December. doi: 10.1109/CVPR.2016.90.

Z. Zhong, L. Zheng, G. Kang, S. Li, and Y. Yang, “Random erasing data augmentation,” 2020. doi: 10.1609/aaai.v34i07.7000.

P. Chen, S. Liu, H. Zhao, and J. Jia, “GridMask Data Augmentation,” Jan. 2020.

S. Zagoruyko and N. Komodakis, “Wide Residual Networks,” in British Machine Vision Conference 2016, BMVC 2016, 2016, vol. 2016-September. doi: 10.5244/C.30.87.

G. Huang, Z. Liu, L. van der Maaten, and K. Q. Weinberger, “Densely connected convolutional networks,” in Proceedings - 30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017, 2017, vol. 2017-January. doi: 10.1109/CVPR.2017.243.

D. Han, J. Kim, and J. Kim, “Deep pyramidal residual networks,” in Proceedings - 30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017, 2017, vol. 2017-January. doi: 10.1109/CVPR.2017.668.




DOI: http://dx.doi.org/10.30700/jst.v13i1.1274

Article Metrics

Abstract view : 256 times
PDF (Indonesian) - 247 times

Refbacks

  • There are currently no refbacks.


Copyright (c) 2023 SISFOTENIKA

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.

Badan Pengelola Jurnal Ilmiah Sistem Informasi dan Teknik Informatika (SISFOTENIKA) STMIK Pontianak.

 

Jurnal Ilmiah SISFOTENIKA terindex di :


   

   

  

    

    

    

   

 

 

 

ISSN Printed : 2087-7897

ISSN Online : 2460-5344


SERTIFIKAT PENGHARGAAN :

Jurnal Ilmiah SISFOTENIKA Terakreditasi Peringkat Empat

 

Partners & Co-Organizers:




Lisensi Creative Commons

Jurnal Ilmiah SISFOTENIKA: STMIK Pontianak Online Journal ISSN Printed (2087-7897) - ISSN Online (2460-5344) licensed under a Lisensi Creative Commons Atribusi 4.0 Internasional. Flag Counter

View My Stats>