CS297 Proposal
Faking Sensor Noise Information.
Justin Chang (justin.h.chang@sjsu.edu)
Advisor: Dr. Chris Pollett
Description:
Pictures taken by digital cameras leave unique imprints on the images they produce. These imprints can be interpreted as noise and be passed through a filter. The goal of this project is to identify a noise filter and camera model pair and impose a noise pattern onto an image that is from a camera different than the camera that the image was originally taken from. This resulting image would be able to fool camera model detection networks of modern benchmarks.
Schedule:
Week 1:
Aug 24 - Aug 31 | Search for Research Papers and work on Proposal |
Week 2:
Aug 31 - Sept 7 | Work on paper [1] |
Week 3:
Sept 7 - Sept 14 | Work on Del 1 |
Week 4:
Sept 14 - Sept 21 | Finish work on Del 1 |
Week 5:
Sept 21 - Sept 28 | Begin work on Del 2 |
Week 6:
Sept 28 - Oct 5 | Continue to work on Del 2 and work on paper [2] |
Week 7:
Oct 5 - Oct 12 | Continue to work on Del 2 |
Week 8:
Oct 12 - Oct 19 | Finish up Del 2 |
Week 9:
Oct 19 - Oct 26 | Begin work on Del 3 |
Week 10:
Oct 26 - Nov 2 | Finish work on Del 3 |
Week 11:
Nov 2 - Nov 9 | Begin work on Del 4 and work on paper [3] |
Week 12:
Nov 9 - Nov 16 | Continue to work on Del 4 and work on paper [4] |
Week 13:
Nov 16 - Nov 23 | Finish work on Del 4 |
Week 14:
Nov 23 - Nov 30 | Begin work on Del 5 and work on paper [5] |
Week 15:
Nov 30 - Dec 7 | Finish Del 5 and work on Final Report |
Deliverables:
The full project will be done when CS298 is completed. The following will
be done by the end of CS297:
1. Find dataset with images and labels with camera model names. Find sensor noise pattern for my own phone based on averages
2. Find sensor noise pattern for my own phone using wavelet transforms.
3. Look at existing camera model detection networks and test them against dataset.
4. Implement a GAN network.
5. Finish report.
References:
[1] J. Lukas, J. Fridrich and M. Goljan, "Digital camera identification from sensor pattern noise," in IEEE Transactions on Information Forensics and Security, vol. 1, no. 2, pp. 205-214, June 2006, doi: 10.1109/TIFS.2006.873602.
[2] D. Cozzolino and L. Verdoliva, "Noiseprint: A CNN-Based Camera Model Fingerprint," in IEEE Transactions on Information Forensics and Security, vol. 15, pp. 144-159, 2020, doi: 10.1109/TIFS.2019.2916364.
[3] S. Samaras, V. Mygdalis and I. Pitas, "Robustness in blind camera identification," 2016 23rd International Conference on Pattern Recognition (ICPR), 2016, pp. 3874-3879, doi: 10.1109/ICPR.2016.7900239.
[4] L. Bondi, L. Baroffio, D. Güera, P. Bestagini, E. J. Delp and S. Tubaro, "First Steps Toward Camera Model Identification With Convolutional Neural Networks," in IEEE Signal Processing Letters, vol. 24, no. 3, pp. 259-263, March 2017, doi: 10.1109/LSP.2016.2641006.
[5] R. Li, Y. Guan and C. Li, "PCA-based denoising of Sensor Pattern Noise for source camera identification," 2014 IEEE China Summit & International Conference on Signal and Information Processing (ChinaSIP), 2014, pp. 436-440, doi: 10.1109/ChinaSIP.2014.6889280.
|