image forensics of high dynamic range imaging · image forensics of high dynamic range imaging 10th...

38
Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored by an EPSRC/Charteris CASE Award P. J. Bateman , A. T. S. Ho, and J. A. Briffa

Upload: others

Post on 15-Mar-2020

9 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Image Forensics of High Dynamic Range Imaging · Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored

Image Forensics of High Dynamic Range Imaging

10th International Workshop on Digital-Forensics & Watermarking

This research is sponsored by an EPSRC/Charteris CASE Award

P. J. Bateman, A. T. S. Ho, and J. A. Briffa

Page 2: Image Forensics of High Dynamic Range Imaging · Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored

Image Forensics

Uncovering facts about an image without actively injecting data

Verify the integrity of a digital image

Source Classification

Camera Identification Processing History Recovery

Forgery Detection

Anomaly Investigation

M. Chen, J. Fridrich, M. Goljan, and J. Lukás, “Image Origin and Integrity Using Sensor Noise,” IEEE Transactions on Information Security and Forensics, 3(1), pp. 74-90, March 2008

H. Farid, “Digital Image Forensics”, American Academy of Forensic Sciences, Washington, DC, 2008

Page 3: Image Forensics of High Dynamic Range Imaging · Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored

Auto-bracketingCamera Merging and Registration

Tone MappingLDR version of HDR Image

HDR Pipeline

Page 4: Image Forensics of High Dynamic Range Imaging · Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored

HDR

-1EV

0EV

+1EV

E. Reinhard, G. Ward, S. Pattanaik, and P. Debevec, “High Dynamic Range Imaging: Acquisition, Display, and Image-Based Lighting,” Morgan Kauffman, ISBN: 978-0-12-585263-0, 2005.

HDR Imaging

Page 5: Image Forensics of High Dynamic Range Imaging · Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored

High Dynamic Range Imaging

Page 6: Image Forensics of High Dynamic Range Imaging · Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored

High Dynamic Range Imaging

Page 7: Image Forensics of High Dynamic Range Imaging · Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored

High Dynamic Range Imaging

Page 8: Image Forensics of High Dynamic Range Imaging · Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored

High Dynamic Range Imaging

Page 9: Image Forensics of High Dynamic Range Imaging · Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored

Image Histogram

LDR

HDR

0

30,000

60,000

0 255

0

350,000

700,000

0 255

High Dynamic Range Imaging

Page 10: Image Forensics of High Dynamic Range Imaging · Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored

HDR is Popular

Page 11: Image Forensics of High Dynamic Range Imaging · Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored

Stats taken from Flickr.com (24-October-2011)

iPhone 4 Camera Useage

Page 12: Image Forensics of High Dynamic Range Imaging · Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored

Can we detect HDR-Processed Images?

Page 13: Image Forensics of High Dynamic Range Imaging · Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored

• An increasingly popular photography method• On-board implementations• More and more HDR images will exist amongst LDR

• EXIF metadata shows little regarding HDR processed images.

• The HDR pipeline differs from manufacturer to manufacturer• Do images contain fingerprints of specific manufacturing pipelines?

• Images can look heavily processed, but are straight off camera• This may fool existing Forensic algorithms

• A novel subject of Image Forensics

Research Motivation

Page 14: Image Forensics of High Dynamic Range Imaging · Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored

Processing History Recovery Anomaly Investigation

Camera Identification

HDR Detection in Image Forensics

Page 15: Image Forensics of High Dynamic Range Imaging · Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored

Auto-bracketingCamera Merging and Registration

LDR version of HDR Image Tone Mapping

HDR Pipeline

Page 16: Image Forensics of High Dynamic Range Imaging · Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored

• Operates on Illuminance-Reflectance model• illuminance can be reduced

• Separate illuminance and reflectance components

I(x,y) = i(x,y) · r(x,y) D = log i(x,y) + log r(x,y)

• (High-Pass) filtering in FFT domain is applied*• Attenuate low frequencies (illuminance)• Preserve high frequencies (reflectance)

*A. V. Oppenheim, R. Schafer, and T. Stockham, “Nonlinear Filtering of Multiplied and Convolved Signals,” in Proceedings of the IEEE, 56(8), pp. 12641291, 1968.

Homomorphic Filtering

Page 17: Image Forensics of High Dynamic Range Imaging · Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored

“Strong” edges contain high and low frequency data

Haloing artefacts are producedG. Qiu, J. Guan, J. Duan, M. Chen, “Tone Mapping for HDR Image using Optimization: A New Closed Form Solution,” 18th International Conference on Pattern Recognition, pp. 996-999, 2006.

The Problem with HF

Page 18: Image Forensics of High Dynamic Range Imaging · Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored

Aim:To accurately classify HDR/LDR images

Device Used:Apple iPhone 4 (Native Camera App)

Method:Capture 100 real-world “landscape” images• 50 HDR• 50 LDR

Images are captured from a tripod to ensure registration processing is minimised

The Experiment

Page 19: Image Forensics of High Dynamic Range Imaging · Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored

HDR Image

Page 20: Image Forensics of High Dynamic Range Imaging · Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored

LDR Image

Page 21: Image Forensics of High Dynamic Range Imaging · Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored

LDR HDR

Spatial Pixel Distribution

Page 22: Image Forensics of High Dynamic Range Imaging · Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored

The Strategy

Read Image(extract luminance)

Canny Edge

Remove Texture

Find “Strongest”

Edge

FFTEdge Data

Classify Edge Data

Majority Voting

Output

Page 23: Image Forensics of High Dynamic Range Imaging · Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored

1. Read Image and Extract Luminance

Page 24: Image Forensics of High Dynamic Range Imaging · Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored

2. Canny Edge Detection

Page 25: Image Forensics of High Dynamic Range Imaging · Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored

3. Threshold Y to B&W

Page 26: Image Forensics of High Dynamic Range Imaging · Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored

4. Morphology: “Open”

Page 27: Image Forensics of High Dynamic Range Imaging · Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored

5. Sobel Edge Detection

Page 28: Image Forensics of High Dynamic Range Imaging · Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored

6. Morphology: “Open”

Page 29: Image Forensics of High Dynamic Range Imaging · Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored

Remove connected edges that do not satisfy:

angle > ±10 of 90°

7. Remove Weaker Edges

Page 30: Image Forensics of High Dynamic Range Imaging · Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored

Remove connected edges that do not satisfy:

angle > ±10 of 90°

7. Remove Weaker Edges

Page 31: Image Forensics of High Dynamic Range Imaging · Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored

8. Plot Pixel Distribution

Page 32: Image Forensics of High Dynamic Range Imaging · Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored

9. Convert to FFT Domain

Page 33: Image Forensics of High Dynamic Range Imaging · Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored

Training Test

No. of images

Edge vectors (per image)

Total no. of feature vectors

90 (45HDR; 45LDR)

10 (5HDR; 5LDR)

100 100

9,000 1,000

2 classes: LDR / HDR

Essentially classifying each edge independently

10. SVM: Train and Classify

Page 34: Image Forensics of High Dynamic Range Imaging · Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored

• Each classification from test set is mapped back to its respective image set (100 per image)

• Majority voting of the results to yield overall image classification

Results

Page 35: Image Forensics of High Dynamic Range Imaging · Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored

• Each classification from test set is mapped back to its respective image set (100 per image)

• Majority voting of the results to yield overall image classification

Test Image Actual Predicted Confidence

1

2

3

4

5

6

7

8

9

10

LDR LDR 87

LDR LDR 92

LDR LDR 100

LDR LDR 91

LDR LDR 90

HDR HDR 88

HDR HDR 99

HDR HDR 80

HDR HDR 69

HDR HDR 55

Results

Page 36: Image Forensics of High Dynamic Range Imaging · Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored

• A proof-of-concept has been presented for detecting HDR processed images

• Halo artefact present in iPhone 4 HDR edges

• Large peak in intensity values characterised at strong edge points

• Strategy for detecting strong edges identified

• Scheme tested and trained on 100 images

• Classification accuracy of 100% (after majority voting)

Summary

Page 37: Image Forensics of High Dynamic Range Imaging · Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored

• Strengthen strategy for detecting strong edges

• Consider edges of all possible orientations

• Greatly increase image set size

• Extend to classify from more device sources

• Classify HDR Apps that created the image

Future Work

Page 38: Image Forensics of High Dynamic Range Imaging · Image Forensics of High Dynamic Range Imaging 10th International Workshop on Digital-Forensics & Watermarking This research is sponsored

Questions?