Research Article
BibTex RIS Cite

Moda Görseli Sınıflandırma: Düzenleyici Teknikler ile Evrişimsel Sinir Ağları Uygulaması

Year 2022, Volume: 15 Issue: 1, 66 - 76, 27.06.2022
https://doi.org/10.54525/tbbmd.1077432

Abstract

Çevrimiçi moda sektörü son yıllarda hızlı bir şekilde büyümektedir. Bu sektörde yer alan moda ürünü görselleri miktarı da sürekli artış göstermektedir. Ürünleri tanımlama ve sınıflandırma yeteneğine sahip bir sistem, görsellere otomatik etiket eklenmesini sağlayarak hızlı erişime olanak verdiği gibi çalışanların iş yükünü de hafifletebilir. Ayrıca moda sınıflandırma sistemi müşterilerin beğenisine dayalı ürünler sunmada kullanılabilir. Büyük miktarlardaki görseli işleyebilmek için ise yüksek performanslı algoritmalara ihtiyaç duyulmaktadır. Son yıllarda derin öğrenme uygulamalarından Evrişimsel Sinir Ağları (CNN) görüntü analizinde başarısı ile ön plana çıkmaktadır. Literatürde bir çok CNN mimarisi yer almakla birlikte, sınıflandırma doğruluğunu arttıracak yeni CNN mimarilerine olan ihtiyaç artan görsel verisi ile birlikte devam etmektedir. Bu çalışma, 10 sınıfa ayrılmış moda ürünü görselleri içeren Fashion-MNIST veri setini kullanarak farklı CNN mimarileri önermektedir. Önerilen mimarilerle amaç L2 ve Dropout düzenleyici yöntemlerin tahmin başarısına olan etkisini araştırmaktır. Bu sayede, verileri daha iyi sınıflandıran CNN modeli araştırılmıştır. Çalışmada önerilen mimariler; temel CNN, L2 düzenleyici ile CNN, Dropout düzenleyici ile CNN ve son olarak her iki düzenleyiciyi içeren CNN modelleridir. Her iki düzenleyici yöntem de ağ ezberlemeyi azaltmıştır. Elde edilen sonuçlara göre Dropout içeren CNN mimarisi %94.3 doğruluk (accuracy) değeri ile en iyi performansı sunan model olmuştur.

References

  • Ö. İnik and E. Ülker, “Derin Öğrenme ve Görüntü Analizinde Kullanılan Derin Öğrenme Modelleri,” Gaziosmanpasa Journal of Scientific Research, vol. 6, no. 3, pp. 85–104, 2017.
  • M. Tripathi, “Analysis of Convolutional Neural Network based Image Classification Techniques,” Journal of Innovative Image Processing, vol. 3, no. 2, pp. 100–117, 2021, doi: 10.36548/jiip.2021.2.003.
  • K. Meshkini, J. Platos, and H. Ghassemain, “An Analysis of Convolutional Neural Network for Fashion Images Classification (Fashion-MNIST),” in International Conference on Intelligent Information Technologies for Industry, 2020, pp. 85–95.
  • Y. Seo and K. shik Shin, “Hierarchical convolutional neural networks for fashion image classification,” Expert Systems with Applications, vol. 116, pp. 328–339, 2019, doi: 10.1016/j.eswa.2018.09.022.
  • Y. Lecun, Y. Bengio, and G. Hinton, “Deep learning,” Nature, vol. 521, no. 7553, pp. 436–444, 2015, doi: 10.1038/nature14539.
  • S. Bhatnagar, D. Ghosal, and M. H. Kolekar, “Classification of fashion article images using convolutional neural networks,” in Fourth International Conference on Image Information Processing, 2017, pp. 357–362, doi: 10.1109/ICIIP.2017.8313740.
  • B. Kolisnik, I. Hogan, and F. Zulkernine, “Condition-CNN: A hierarchical multi-label fashion image classification model,” Expert Systems with Applications, vol. 182, p. 115195, 2021, doi: 10.1016/j.eswa.2021.115195.
  • A. Krizhevsky, I. Sutskever, and G. E. Hinton, “ImageNet Classification with Deep Convolutional Neural Networks,” Handbook of Approximation Algorithms and Metaheuristics, pp. 1–9, 2012, doi: 10.1201/9781420010749.
  • M. Kayed, A. Anter, and H. Mohamed, “Classification of Garments from Fashion MNIST Dataset Using CNN LeNet-5 Architecture,” in International Conference on Innovative Trends in Communication and Computer Engineering, 2020, pp. 238–243, doi: 10.1109/ITCE48509.2020.9047776.
  • A. Orendorff, “10 Trends Styling 2021’s Ecommerce Fashion Industry: Growth + Data in Online Apparel & Accessories Market,” Common Thread Collective, 2022. [Online]. Available: https://commonthreadco.com/blogs/coachs-corner/fashion-ecommerce-industry-trends#fashion-ecommerce-trend-market.
  • H. Cho, C. Ahn, K. M. Yoo, J. Seol, and S. G. Lee, “Leveraging class hierarchy in fashion classification,” in International Conference on Computer Vision Workshop, 2019, pp. 3197–3200, doi: 10.1109/ICCVW.2019.00398.
  • S. G. Eshwar, J. Gautham Ganesh Prabhu, A. V. Rishikesh, N. A. Charan, and V. Umadevi, “Apparel classification using Convolutional Neural Networks,” in International Conference on ICT in Business, Industry, and Government, 2016, pp. 1–5, doi: 10.1109/ICTBIG.2016.7892641.
  • K. Hara, V. Jagadeesh, and R. Piramuthu, “Fashion apparel detection: The role of deep convolutional neural network and pose-dependent priors,” in IEEE Winter Conference on Applications of Computer Vision, 2016, doi: 10.1109/WACV.2016.7477611.
  • M. Anjan and V. Abhishek, “Fashion recommendation system using CNN,” International Journal of Advance Research, Ideas and Innovations in Technology, vol. 6, no. 3, pp. 780–783, 2020.
  • L. Bossard, M. Dantone, C. Leistner, C. Wengert, T. Quack, and L. Van Gool, “Apparel classification with style,” Lecture Notes in Computer Science, pp. 321–335, 2013, doi: 10.1007/978-3-642-37447-0_25.
  • F. Özbilgin and C. Tepe, “Robotik Uygulamalar İçin Derin Öğrenme Tabanlı Nesne Tespiti ve Sınıflandırması,” Karadeniz Fen Bilimleri Dergisi, vol. 10, no. 1, pp. 205–213, 2020, doi: 10.31466/kfbd.734393.
  • M. Mutlu Bilgin, K. Özdem, and M. A. Akcayol, “Derin Öğrenme ile Kuş Türü Sınıflandırma: Karşılaştırmalı Bir Çalışma,” Journal of Polytechnic, 2021, doi: 10.2339/politeknik.904933.
  • B. Ay, “HorrorFace: Derin Öğrenme Tabanlı Korkutucu Yüzlerin Tespiti v e Sınıflandırılması,” Bilişim Teknolojileri Dergisi, vol. 14, no. 4, pp. 435–443, 2021, doi: 10.17671/gazibtd.875816.
  • B. Baheti, S. Gajre, and S. Talbar, “Detection of Driver Distraction Using Convolutional Neural Network,” Proceedings of the IEEE conference on computer vision and pattern recognition workshop, pp. 1032–1038, 2018, doi: 10.1007/978-981-16-4149-7_28.
  • Y. Sun, B. Xue, M. Zhang, and G. G. Yen, “Evolving Deep Convolutional Neural Networks for Image Classification,” IEEE Transactions on Evolutionary Computation, vol. 24, no. 2, pp. 394–407, 2020, doi: 10.1109/TEVC.2019.2916183.
  • K. V. Greeshma and K. Sreekumar, “Hyperparameter optimization and regularization on fashion-MNIST classification,” International Journal of Recent Technology and Engineering, vol. 8, no. 2, pp. 3713–3719, 2019, doi: 10.35940/ijrte.B3092.078219.
  • T. Hur, L. Kim, and D. K. Park, “Quantum convolutional neural network for classical data classification,” arxiv preprint:2108.00661, pp. 1–16, 2021.
  • A. S. Henrique et al., “Classifying Garments from Fashion-MNIST Dataset Through CNNs,” Advances in Science, Technology and Engineering Systems Journal, vol. 6, no. 1, pp. 989–994, 2021, doi: 10.25046/aj0601109.
  • O. M. Khanday, S. Dadvandipour, and M. A. Lone, “Effect of filter sizes on image classification in CNN: A case study on CFIR10 and fashion-MNIST datasets,” International Journal of Artificial Intelligence, vol. 10, no. 4, pp. 872–878, 2021, doi: 10.11591/ijai.v10.i4.pp872-878.
  • H. Xiao, K. Rasul, and R. Vollgraf, “Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms,” arxiv preprint: 1708.07747, pp. 1–6, 2017.
  • I. Goodfellow, Y. Bengio, and A. Courville, Deep Learning. MIT Press, 2016.
  • N. Srivastava, G. Hinton, A. Krizhevsky, I. Sutskever, and R. Salakhutdinov, “Dropout: A Simple Way to Prevent Neural Networks from Overfittin,” Journal of Machine Learning Research 15, pp. 1929–1958, 2014, doi: 10.1016/0370-2693(93)90272-J.
Year 2022, Volume: 15 Issue: 1, 66 - 76, 27.06.2022
https://doi.org/10.54525/tbbmd.1077432

Abstract

References

  • Ö. İnik and E. Ülker, “Derin Öğrenme ve Görüntü Analizinde Kullanılan Derin Öğrenme Modelleri,” Gaziosmanpasa Journal of Scientific Research, vol. 6, no. 3, pp. 85–104, 2017.
  • M. Tripathi, “Analysis of Convolutional Neural Network based Image Classification Techniques,” Journal of Innovative Image Processing, vol. 3, no. 2, pp. 100–117, 2021, doi: 10.36548/jiip.2021.2.003.
  • K. Meshkini, J. Platos, and H. Ghassemain, “An Analysis of Convolutional Neural Network for Fashion Images Classification (Fashion-MNIST),” in International Conference on Intelligent Information Technologies for Industry, 2020, pp. 85–95.
  • Y. Seo and K. shik Shin, “Hierarchical convolutional neural networks for fashion image classification,” Expert Systems with Applications, vol. 116, pp. 328–339, 2019, doi: 10.1016/j.eswa.2018.09.022.
  • Y. Lecun, Y. Bengio, and G. Hinton, “Deep learning,” Nature, vol. 521, no. 7553, pp. 436–444, 2015, doi: 10.1038/nature14539.
  • S. Bhatnagar, D. Ghosal, and M. H. Kolekar, “Classification of fashion article images using convolutional neural networks,” in Fourth International Conference on Image Information Processing, 2017, pp. 357–362, doi: 10.1109/ICIIP.2017.8313740.
  • B. Kolisnik, I. Hogan, and F. Zulkernine, “Condition-CNN: A hierarchical multi-label fashion image classification model,” Expert Systems with Applications, vol. 182, p. 115195, 2021, doi: 10.1016/j.eswa.2021.115195.
  • A. Krizhevsky, I. Sutskever, and G. E. Hinton, “ImageNet Classification with Deep Convolutional Neural Networks,” Handbook of Approximation Algorithms and Metaheuristics, pp. 1–9, 2012, doi: 10.1201/9781420010749.
  • M. Kayed, A. Anter, and H. Mohamed, “Classification of Garments from Fashion MNIST Dataset Using CNN LeNet-5 Architecture,” in International Conference on Innovative Trends in Communication and Computer Engineering, 2020, pp. 238–243, doi: 10.1109/ITCE48509.2020.9047776.
  • A. Orendorff, “10 Trends Styling 2021’s Ecommerce Fashion Industry: Growth + Data in Online Apparel & Accessories Market,” Common Thread Collective, 2022. [Online]. Available: https://commonthreadco.com/blogs/coachs-corner/fashion-ecommerce-industry-trends#fashion-ecommerce-trend-market.
  • H. Cho, C. Ahn, K. M. Yoo, J. Seol, and S. G. Lee, “Leveraging class hierarchy in fashion classification,” in International Conference on Computer Vision Workshop, 2019, pp. 3197–3200, doi: 10.1109/ICCVW.2019.00398.
  • S. G. Eshwar, J. Gautham Ganesh Prabhu, A. V. Rishikesh, N. A. Charan, and V. Umadevi, “Apparel classification using Convolutional Neural Networks,” in International Conference on ICT in Business, Industry, and Government, 2016, pp. 1–5, doi: 10.1109/ICTBIG.2016.7892641.
  • K. Hara, V. Jagadeesh, and R. Piramuthu, “Fashion apparel detection: The role of deep convolutional neural network and pose-dependent priors,” in IEEE Winter Conference on Applications of Computer Vision, 2016, doi: 10.1109/WACV.2016.7477611.
  • M. Anjan and V. Abhishek, “Fashion recommendation system using CNN,” International Journal of Advance Research, Ideas and Innovations in Technology, vol. 6, no. 3, pp. 780–783, 2020.
  • L. Bossard, M. Dantone, C. Leistner, C. Wengert, T. Quack, and L. Van Gool, “Apparel classification with style,” Lecture Notes in Computer Science, pp. 321–335, 2013, doi: 10.1007/978-3-642-37447-0_25.
  • F. Özbilgin and C. Tepe, “Robotik Uygulamalar İçin Derin Öğrenme Tabanlı Nesne Tespiti ve Sınıflandırması,” Karadeniz Fen Bilimleri Dergisi, vol. 10, no. 1, pp. 205–213, 2020, doi: 10.31466/kfbd.734393.
  • M. Mutlu Bilgin, K. Özdem, and M. A. Akcayol, “Derin Öğrenme ile Kuş Türü Sınıflandırma: Karşılaştırmalı Bir Çalışma,” Journal of Polytechnic, 2021, doi: 10.2339/politeknik.904933.
  • B. Ay, “HorrorFace: Derin Öğrenme Tabanlı Korkutucu Yüzlerin Tespiti v e Sınıflandırılması,” Bilişim Teknolojileri Dergisi, vol. 14, no. 4, pp. 435–443, 2021, doi: 10.17671/gazibtd.875816.
  • B. Baheti, S. Gajre, and S. Talbar, “Detection of Driver Distraction Using Convolutional Neural Network,” Proceedings of the IEEE conference on computer vision and pattern recognition workshop, pp. 1032–1038, 2018, doi: 10.1007/978-981-16-4149-7_28.
  • Y. Sun, B. Xue, M. Zhang, and G. G. Yen, “Evolving Deep Convolutional Neural Networks for Image Classification,” IEEE Transactions on Evolutionary Computation, vol. 24, no. 2, pp. 394–407, 2020, doi: 10.1109/TEVC.2019.2916183.
  • K. V. Greeshma and K. Sreekumar, “Hyperparameter optimization and regularization on fashion-MNIST classification,” International Journal of Recent Technology and Engineering, vol. 8, no. 2, pp. 3713–3719, 2019, doi: 10.35940/ijrte.B3092.078219.
  • T. Hur, L. Kim, and D. K. Park, “Quantum convolutional neural network for classical data classification,” arxiv preprint:2108.00661, pp. 1–16, 2021.
  • A. S. Henrique et al., “Classifying Garments from Fashion-MNIST Dataset Through CNNs,” Advances in Science, Technology and Engineering Systems Journal, vol. 6, no. 1, pp. 989–994, 2021, doi: 10.25046/aj0601109.
  • O. M. Khanday, S. Dadvandipour, and M. A. Lone, “Effect of filter sizes on image classification in CNN: A case study on CFIR10 and fashion-MNIST datasets,” International Journal of Artificial Intelligence, vol. 10, no. 4, pp. 872–878, 2021, doi: 10.11591/ijai.v10.i4.pp872-878.
  • H. Xiao, K. Rasul, and R. Vollgraf, “Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms,” arxiv preprint: 1708.07747, pp. 1–6, 2017.
  • I. Goodfellow, Y. Bengio, and A. Courville, Deep Learning. MIT Press, 2016.
  • N. Srivastava, G. Hinton, A. Krizhevsky, I. Sutskever, and R. Salakhutdinov, “Dropout: A Simple Way to Prevent Neural Networks from Overfittin,” Journal of Machine Learning Research 15, pp. 1929–1958, 2014, doi: 10.1016/0370-2693(93)90272-J.
There are 27 citations in total.

Details

Primary Language Turkish
Subjects Engineering
Journal Section Makaleler(Araştırma)
Authors

Şule Öztürk Birim 0000-0001-7544-8588

Early Pub Date June 27, 2022
Publication Date June 27, 2022
Published in Issue Year 2022 Volume: 15 Issue: 1

Cite

APA Öztürk Birim, Ş. (2022). Moda Görseli Sınıflandırma: Düzenleyici Teknikler ile Evrişimsel Sinir Ağları Uygulaması. Türkiye Bilişim Vakfı Bilgisayar Bilimleri Ve Mühendisliği Dergisi, 15(1), 66-76. https://doi.org/10.54525/tbbmd.1077432
AMA Öztürk Birim Ş. Moda Görseli Sınıflandırma: Düzenleyici Teknikler ile Evrişimsel Sinir Ağları Uygulaması. TBV-BBMD. June 2022;15(1):66-76. doi:10.54525/tbbmd.1077432
Chicago Öztürk Birim, Şule. “Moda Görseli Sınıflandırma: Düzenleyici Teknikler Ile Evrişimsel Sinir Ağları Uygulaması”. Türkiye Bilişim Vakfı Bilgisayar Bilimleri Ve Mühendisliği Dergisi 15, no. 1 (June 2022): 66-76. https://doi.org/10.54525/tbbmd.1077432.
EndNote Öztürk Birim Ş (June 1, 2022) Moda Görseli Sınıflandırma: Düzenleyici Teknikler ile Evrişimsel Sinir Ağları Uygulaması. Türkiye Bilişim Vakfı Bilgisayar Bilimleri ve Mühendisliği Dergisi 15 1 66–76.
IEEE Ş. Öztürk Birim, “Moda Görseli Sınıflandırma: Düzenleyici Teknikler ile Evrişimsel Sinir Ağları Uygulaması”, TBV-BBMD, vol. 15, no. 1, pp. 66–76, 2022, doi: 10.54525/tbbmd.1077432.
ISNAD Öztürk Birim, Şule. “Moda Görseli Sınıflandırma: Düzenleyici Teknikler Ile Evrişimsel Sinir Ağları Uygulaması”. Türkiye Bilişim Vakfı Bilgisayar Bilimleri ve Mühendisliği Dergisi 15/1 (June 2022), 66-76. https://doi.org/10.54525/tbbmd.1077432.
JAMA Öztürk Birim Ş. Moda Görseli Sınıflandırma: Düzenleyici Teknikler ile Evrişimsel Sinir Ağları Uygulaması. TBV-BBMD. 2022;15:66–76.
MLA Öztürk Birim, Şule. “Moda Görseli Sınıflandırma: Düzenleyici Teknikler Ile Evrişimsel Sinir Ağları Uygulaması”. Türkiye Bilişim Vakfı Bilgisayar Bilimleri Ve Mühendisliği Dergisi, vol. 15, no. 1, 2022, pp. 66-76, doi:10.54525/tbbmd.1077432.
Vancouver Öztürk Birim Ş. Moda Görseli Sınıflandırma: Düzenleyici Teknikler ile Evrişimsel Sinir Ağları Uygulaması. TBV-BBMD. 2022;15(1):66-7.

Article Acceptance

Use user registration/login to upload articles online.

The acceptance process of the articles sent to the journal consists of the following stages:

1. Each submitted article is sent to at least two referees at the first stage.

2. Referee appointments are made by the journal editors. There are approximately 200 referees in the referee pool of the journal and these referees are classified according to their areas of interest. Each referee is sent an article on the subject he is interested in. The selection of the arbitrator is done in a way that does not cause any conflict of interest.

3. In the articles sent to the referees, the names of the authors are closed.

4. Referees are explained how to evaluate an article and are asked to fill in the evaluation form shown below.

5. The articles in which two referees give positive opinion are subjected to similarity review by the editors. The similarity in the articles is expected to be less than 25%.

6. A paper that has passed all stages is reviewed by the editor in terms of language and presentation, and necessary corrections and improvements are made. If necessary, the authors are notified of the situation.

0

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