The Unseen Challenge Data Sets

1
The Unseen Challenge Data Sets

• Anderson Rocha Walter Scheirer
• Siome Goldenstein Terrance Boult

2
The Data Sets

• Two data sets are provided
• PNG lossless compression
• JPEG lossy compression
• Prevalence of images on the Internet
• Sources Google images, Yahoo Images, and Flickr

3
Message Sizes

• For each tool, we provide four different
  embedding size
• Tiny lt 5 of the channel capacity
• Small gt 5 lt 15 of the channel capacity
• Medium gt 15 lt 40 of the channel capacity
• Large gt 40 of the channel capacity
• For the PNG set, the message size is explicitly
  stated
• For the JPEG set, the message size is NOT stated

4
Message Content

• Random bit sequences
• Snippets of mp3 songs
• Plain text
• Other images

A B C
5
Categories

• Each set consists of clean and stego images
• Clean set
• Modified cropping, overlay, object-appending
• Non-modified original
• Stego set
• 4 categories for JPEG, 3 categories for PNG, one
  for each tool

6
Categories

• JPEG subcategories
• Stego
• Animals
• Business
• Maps
• Natural
• Tourist
• Vacation
• Clean
• Misc

7
Clean Manipulated Images
Object Appending
Image Cropping
Overlay
8
PNG Tools

• Camaleão (http//www.ic.unicamp.br/rocha/sci/steg
  o)
• Simple LSB insertion/modification software
• Uses cyclic permutations and block ciphering to
  hide messages in LSBs
• SecurEngine
• (http//www.sharewareplaza.com/SecurEngine-dow
  nload_4268.html)
• Incorporates 5 crypto algorithms Blowfish, Gost,
  Vernam, Cast256, and Mars
• LSB encoding

9
PNG Tools

• Stash-It
• (http//www.smalleranimals.com/stash.htm)
• Windows based stego tool
• Simple LSB insertion/modification software
• No encryption feature

10
JPEG Tools

• F5
• (http//www.inf.tu-dresden.de/aw4)
• Resilient to ?2 statistical attack
• Instead of replacing LSBs directly, F5 decreases
  the absolute value of the DCT coefficients
• Chooses DCT coefficients randomly
• Matrix embedding
• JPHide
• (http//linux01.gwdg.de/alatham)
• Uses blowfish to generate a stream of
  pseudo-random control bits to define bit
  encodings
• Large embeddings trivial to detect

11
JPEG Tools

• JSteg
• (http//zooid.org/paul/crypto/jsteg)
• 40 bit RC4 Encryption
• Channel capacity determination
• LSB encoding in quantized DCT coefficients
• Outguess
• (http//www.outguess.org/detection.php)
• Preserves statistics based on frequency counts
• Seed based iterator available to choose embedding
  locations
• Change minimization calculation for each seed
• Remains one of the most difficult tools to detect

12
PNG Data Set - Breakdown

• Training

4,000 total images in the PNG clean category
4,731 total images in the PNG stego category
13
PNG Data Set - Breakdown

• Testing

2,993 total images in the PNG stego category
14
JPEG Data Set - Breakdown

• Training

29,185 total images in the JPEG stego category
15
JPEG Data Set - Breakdown

• Training

29,185 total images in the JPEG stego category
16
JPEG Data Set - Breakdown

• Testing

4,596 total images in the JPEG stego category
17
Sample Usage stegdetect

• JPEG Training Set

Detected, C correct algorithm detected
Detected, I incorrect algorithm detected
Overall false detect rate for the clean image set
is 8.6
18
Sample Usage stegdetect

• JPEG Testing Set

Overall false detect rate for the clean image set
is 8.0
19
Sample Usage stegdetect

• Detailed results for JPHide Test Set

20
Sample Usage stegdetect

• Conclusions
• Significant differences between the results of
  training and testing
• Weaker performance overall for testing
• Designed difficulty of testing set
• Stegdetect performs poorly for large embeddings
  (non-intuitive), as well as small and tiny
  embeddings (expected)

21
The Unseen Challenge Data Sets

• Lossy (JPEG) and Lossless (PNG) imagery
• 3 tools for PNG set, 4 tools for JPEG set
• 4 distinct embedding sizes for PNG, varying sizes
  for JPEG
• Clean imagery across all sets

22
The Unseen Challenge Data Sets

• Valid approaches for use
• Detection
• Detection and recovery (size or content)
• Detection and destruction
• Fusion

No standard data set exists for steg evaluation!
This set is a step in that direction!
23
Download!

• http//www.liv.ic.unicamp.br/wvu/datasets.php