Advertisement

Camera Model Identification Using Transfer Learning

  • Mohit KulkarniEmail author
  • Shivam Kakad
  • Rahul Mehra
  • Bhavya Mehta
Conference paper
  • 13 Downloads
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 1101)

Abstract

Wide series of forensic problems can be solved by detecting the camera model from copyright infringement to ownership attribution. There are many proposed methods for detection of the camera model. A method to identify the camera model of any image is proposed in this paper. It involved feature extraction and classification. CNN-based architectures are best suited for the task of image classification.

Keywords

Camera model identification Convolutional neural networks ResNet Transfer learning 

Notes

Acknowledgements

The authors feel grateful to and they wish their profound indebtedness to their guide Prof. Milind Kamble, Department of Electronics and Telecommunication, Vishwakarma Institute of Technology, Pune. The authors also express their gratitude to Prof. Dr. R. M. Jalnekar, Director, and Prof. Dr. Shripad Bhatlawande, Head, Department of Electronics and Telecommunication, for their help in completion of the project. The authors also thank all the anonymous reviewers of this paper whose comments helped to improve the paper.

References

  1. 1.
    Tuama, Amel, Frédéric Comby, Marc Chaumont. 2016. Camera model identification with the use of deep convolutional neural networks. Available via http://www.lirmm.fr/~chaumont/publications/WIFS-2016_TUAMA_COMBY_CHAUMONT_Camera_Model_Identification_With_CNN_slides.pdf. Accessed January 13, 2018.
  2. 2.
    Antorsae. 2018. IEEE’s Signal Processing Society—Camera Model Identification Kaggle Competition. Available via https://github.com/antorsae/sp-society-camera-model-identification. Accessed March 25, 2018.
  3. 3.
    Shah, Anuj. 2018. Transfer learning using Keras. Available via https://github.com/anujshah1003/Transfer-Learning-in-keras—custom-data. Accessed February 3, 2018.
  4. 4.
    Anonymous. 2018. Convolutional neural networks for visual recognition. Available via http://cs231n.github.io/convolutional-networks/. Accessed February 21, 2018.
  5. 5.
    Bondi, Luca, Luca Baroffio, David Guera, Paolo Bestagini, Edward Delp, and Stefano Tubaro. 2015 First steps toward camera model identification with convolutional neural networks. Journal of Latex Class Files 14: 1–4. arXiv:1603.01068.
  6. 6.
    Ujjwalkarn. 2016. An intuitive explanation of convolutional neural networks. Available via https://ujjwalkarn.me/2016/08/11/intuitive-explanation-convnets/. Accessed March 17, 2018.
  7. 7.
    Kuzin, Artur, Artur Fattakhov, Ilya Kibardin, Vladimir Iglovikov, and Ruslan Dautov. 2018. Camera model identification using convolutional neural networks, 1–4. arXiv:1810.02981.
  8. 8.
    Jay, Prakash. 2017. Transfer learning using Keras. Available via https://medium.com/@14prakash/transfer-learning-using-keras-d804b2e04ef8. Accessed February 21, 2018.
  9. 9.
    Kaggle. 2018. ResNet-50 Pre-trained Model for Keras. Available via https://www.kaggle.com/keras/resnet50. Accessed February 26, 2018.
  10. 10.
    Google. 2015. Tensorflow. Available via https://www.tensorflow.org/. Accessed January, February 9, 2018.
  11. 11.
    Koustubh. 2018. ResNet, AlexNet, VGGNet, inception: Understanding various architectures of convolutional networks. Available via http://cv-tricks.com/cnn/understand-resnet-alexnet-vgg-inception/. Accessed March 23 2018.
  12. 12.
    He, K, X. Zhang, S. Ren, and J. Sun. 2015. Deep residual learning for image recognition, 1–8. arXiv:1512.03385.

Copyright information

© Springer Nature Singapore Pte Ltd. 2020

Authors and Affiliations

  • Mohit Kulkarni
    • 1
    Email author
  • Shivam Kakad
    • 1
  • Rahul Mehra
    • 1
  • Bhavya Mehta
    • 1
  1. 1.Electronics and Telecommunication DepartmentVishwakarma Institute of TechnologyPuneIndia

Personalised recommendations