VGG16 İkili Sınıflandırıcı Derin Sinir Ağının Gürültülü ve Bulanık Görüntüler için Değerlendirilmesi
Öz
Derin öğrenme ağları, görüntü sınıflandırma uygulamaları için önemli bir araç haline gelmiştir. Görüntülerdeki bozulmalar, sınıflandırıcının performansının önemli ölçüde düşmesine neden olabilir. Bu makalede, bozuk girişler altında VGG16 ağının ikili sınıflandırma performansı için karşılaştırmalı bir araştırma sunulmuştur. Bu amaçla, çeşitli seviyelerde bozulmuş görüntüler ve Gauss gürültüsü, Tuz ve Biber gürültüsü ve bulanıklık efekti ile sabit seviyelerde görüntüler test için kullanılmıştır. VGG16'nın evrişimli katmanları, son üç evrişimli katman ve sınıflandırma için yoğun bir katman hariç dondurulmuştur. Deneysel sonuçlara göre, bozulmanın etkisi arttıkça, derin öğrenme sınıflandırıcısının performansı önemli ölçüde düşmektedir. Bozulma etkilerini içeren artırılmış eğitim durumunda, sonuçlar önemli ölçüde iyileştirilmiştir.
Kaynakça
- A. Voulodimos, N. Doulamis, A. Doulamis, and E. Protopapadakis, “Deep Learning for Computer Vision: A Brief Review,” Computational Intelligence and Neuroscience, vol. 2018. Hindawi Limited, 2018.
- T. Young, D. Hazarika, S. Poria, and E. Cambria, “Recent trends in deep learning based natural language processing,” ieee Comput. Intell. Mag., vol. 13, no. 3, pp. 55–75, 2018.
- H. Purwins, B. Li, T. Virtanen, J. Schlüter, S.-Y. Chang, and T. Sainath, “Deep learning for audio signal processing,” IEEE J. Sel. Top. Signal Process., vol. 13, no. 2, pp. 206–219, 2019.
- F. Altaf, S. M. S. Islam, N. Akhtar, and N. K. Janjua, “Going Deep in Medical Image Analysis: Concepts, Methods, Challenges, and Future Directions,” IEEE Access, vol. 7, pp. 99540–99572, 2019.
- F. Chollet, “Keras,” GitHub repository. GitHub, 2015.
- M. Abadi et al., “TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems,” arXiv Prepr. arXiv1603.04467, 2016.
- D. Yu et al., “An introduction to computational networks and the computational network toolkit,” 2014.
- R. Al-Rfou et al., “Theano: A {Python} framework for fast computation of mathematical expressions,” arXiv e-prints, vol. abs/1605.0, May 2016.
- Y. LeCun et al., “Handwritten digit recognition with a back-propagation network,” in Advances in neural information processing systems, 1990, pp. 396–404.
- C. Shorten and T. M. Khoshgoftaar, “A survey on Image Data Augmentation for Deep Learning,” J. Big Data, vol. 6, no. 1, p. 60, 2019.
- S. Dodge and L. Karam, “Understanding how image quality affects deep neural networks,” in 2016 eighth international conference on quality of multimedia experience (QoMEX), 2016, pp. 1–6.
- S. Dodge and L. Karam, “A study and comparison of human and deep learning recognition performance under visual distortions,” in 2017 26th international conference on computer communication and networks (ICCCN), 2017, pp. 1–7.
- I. Vasiljevic, A. Chakrabarti, and G. Shakhnarovich, “Examining the impact of blur on recognition by convolutional networks,” arXiv Prepr. arXiv1611.05760, 2016.
- D. Yin, R. G. Lopes, J. Shlens, E. D. Cubuk, and J. Gilmer, “A fourier perspective on model robustness in computer vision,” in Advances in Neural Information Processing Systems, 2019, pp. 13255–13265.
- E. Rusak et al., “Increasing the robustness of DNNs against image corruptions by playing the Game of Noise,” arXiv Prepr. arXiv2001.06057, 2020.
- C. Kamann and C. Rother, “Benchmarking the robustness of semantic segmentation models with respect to common corruptions,” Int. J. Comput. Vis., pp. 1–22, 2020.
- K. Simonyan and A. Zisserman, “Very deep convolutional networks for large-scale image recognition,” arXiv Prepr. arXiv1409.1556, 2014.
Öz
Deep learning networks has become an important tool for image classification applications. Distortions on images may cause the performance of a classifier to decrease significantly. In the present paper, a comparative investigation for binary classification performance of VGG16 network under corrupted inputs has been presented. For this purpose, images corrupted at various levels and fixed levels with Gaussian noise, Salt and Pepper noise and blur effect were used for testing. Convolutional layers of the VGG16 were frozen except the last three convolutional layers and a dense layer for binary classification was added. According to experimental results, as the effect of distortion is increased, performance of the deep learning classifier drops significantly. In the case of augmented training with distortion effects, the results were improved significantly.
Anahtar Kelimeler
deep neural network VGG16 pretrained networks fine tuning augmentation corrupted images
Kaynakça
- A. Voulodimos, N. Doulamis, A. Doulamis, and E. Protopapadakis, “Deep Learning for Computer Vision: A Brief Review,” Computational Intelligence and Neuroscience, vol. 2018. Hindawi Limited, 2018.
- T. Young, D. Hazarika, S. Poria, and E. Cambria, “Recent trends in deep learning based natural language processing,” ieee Comput. Intell. Mag., vol. 13, no. 3, pp. 55–75, 2018.
- H. Purwins, B. Li, T. Virtanen, J. Schlüter, S.-Y. Chang, and T. Sainath, “Deep learning for audio signal processing,” IEEE J. Sel. Top. Signal Process., vol. 13, no. 2, pp. 206–219, 2019.
- F. Altaf, S. M. S. Islam, N. Akhtar, and N. K. Janjua, “Going Deep in Medical Image Analysis: Concepts, Methods, Challenges, and Future Directions,” IEEE Access, vol. 7, pp. 99540–99572, 2019.
- F. Chollet, “Keras,” GitHub repository. GitHub, 2015.
- M. Abadi et al., “TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems,” arXiv Prepr. arXiv1603.04467, 2016.
- D. Yu et al., “An introduction to computational networks and the computational network toolkit,” 2014.
- R. Al-Rfou et al., “Theano: A {Python} framework for fast computation of mathematical expressions,” arXiv e-prints, vol. abs/1605.0, May 2016.
- Y. LeCun et al., “Handwritten digit recognition with a back-propagation network,” in Advances in neural information processing systems, 1990, pp. 396–404.
- C. Shorten and T. M. Khoshgoftaar, “A survey on Image Data Augmentation for Deep Learning,” J. Big Data, vol. 6, no. 1, p. 60, 2019.
- S. Dodge and L. Karam, “Understanding how image quality affects deep neural networks,” in 2016 eighth international conference on quality of multimedia experience (QoMEX), 2016, pp. 1–6.
- S. Dodge and L. Karam, “A study and comparison of human and deep learning recognition performance under visual distortions,” in 2017 26th international conference on computer communication and networks (ICCCN), 2017, pp. 1–7.
- I. Vasiljevic, A. Chakrabarti, and G. Shakhnarovich, “Examining the impact of blur on recognition by convolutional networks,” arXiv Prepr. arXiv1611.05760, 2016.
- D. Yin, R. G. Lopes, J. Shlens, E. D. Cubuk, and J. Gilmer, “A fourier perspective on model robustness in computer vision,” in Advances in Neural Information Processing Systems, 2019, pp. 13255–13265.
- E. Rusak et al., “Increasing the robustness of DNNs against image corruptions by playing the Game of Noise,” arXiv Prepr. arXiv2001.06057, 2020.
- C. Kamann and C. Rother, “Benchmarking the robustness of semantic segmentation models with respect to common corruptions,” Int. J. Comput. Vis., pp. 1–22, 2020.
- K. Simonyan and A. Zisserman, “Very deep convolutional networks for large-scale image recognition,” arXiv Prepr. arXiv1409.1556, 2014.
Ayrıntılar
| Birincil Dil | İngilizce |
|---|---|
| Konular | Yapay Zeka |
| Bölüm | Makaleler |
| Yazarlar | |
| Yayımlanma Tarihi | 30 Aralık 2020 |
| Gönderilme Tarihi | 22 Nisan 2020 |
| Kabul Tarihi | 2 Aralık 2020 |
| Yayımlandığı Sayı | Yıl 2020Cilt: 3 Sayı: 3 |
Cited By
Deep Neural Networks Based on Transfer Learning Approaches to Classification of Gun and Knife Images
Sakarya University Journal of Computer and Information Sciences
https://doi.org/10.35377/saucis.04.01.891308
LSTM tabanlı Derin Sinir Ağı ile Beşinci Nesil Küçük Hücre Ağlarında El Değiştirme Tahmini
Bilecik Şeyh Edebali Üniversitesi Fen Bilimleri Dergisi
https://doi.org/10.35193/bseufbd.840927
The papers in this journal are licensed under a Creative Commons Attribution-NonCommercial 4.0 International License