Presentation is loading. Please wait.

Presentation is loading. Please wait.

IEEE-WVU, Anchorage - 2008  1 The Unseen Challenge Data Sets Anderson Rocha Walter Scheirer Siome Goldenstein Terrance Boult.

Similar presentations


Presentation on theme: "IEEE-WVU, Anchorage - 2008  1 The Unseen Challenge Data Sets Anderson Rocha Walter Scheirer Siome Goldenstein Terrance Boult."— Presentation transcript:

1 IEEE-WVU, Anchorage - 2008  1 The Unseen Challenge Data Sets Anderson Rocha Walter Scheirer Siome Goldenstein Terrance Boult

2 IEEE-WVU, Anchorage - 2008  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 IEEE-WVU, Anchorage - 2008  3 Message Sizes For each tool, we provide four different embedding size: –Tiny: < 5% of the channel capacity –Small: > 5% & < 15% of the channel capacity –Medium: > 15% & < 40% of the channel capacity –Large: > 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 IEEE-WVU, Anchorage - 2008  4 Message Content Random bit sequences Snippets of mp3 songs Plain text Other images A B C

5 IEEE-WVU, Anchorage - 2008  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 IEEE-WVU, Anchorage - 2008  6 Categories JPEG subcategories –Stego Animals Business Maps Natural Tourist Vacation –Clean Misc

7 IEEE-WVU, Anchorage - 2008  7 Clean Manipulated Images Object Appending Image Cropping Overlay

8 IEEE-WVU, Anchorage - 2008  8 PNG Tools Camaleão (http://www.ic.unicamp.br/~rocha/sci/stego)http://www.ic.unicamp.br/~rocha/sci/stego –Simple LSB insertion/modification software –Uses cyclic permutations and block ciphering to hide messages in LSBs SecurEngine (http://www.sharewareplaza.com/SecurEngine- download_4268.html)http://www.sharewareplaza.com/SecurEngine- download_4268.html –Incorporates 5 crypto algorithms: Blowfish, Gost, Vernam, Cast256, and Mars –LSB encoding

9 IEEE-WVU, Anchorage - 2008  9 PNG Tools Stash-It (http://www.smalleranimals.com/stash.htm)http://www.smalleranimals.com/stash.htm –Windows based stego tool –Simple LSB insertion/modification software –No encryption feature

10 IEEE-WVU, Anchorage - 2008  10 JPEG Tools F5 (http://www.inf.tu-dresden.de/~aw4)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)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 IEEE-WVU, Anchorage - 2008  11 JPEG Tools JSteg (http://zooid.org/~paul/crypto/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)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 IEEE-WVU, Anchorage - 2008  12 PNG Data Set - Breakdown Training TinySmallMediumLarge Camaleão400 SecurEngine380387385380 Stash-It399400 Total1,1791,1871,1851,180 Non- Modified 2,000 Append- Modified 666 Crop- modified 667 Overlay- modified 667 Total4,000 4,731 total images in the PNG stego category 4,000 total images in the PNG clean category

13 IEEE-WVU, Anchorage - 2008  13 PNG Data Set - Breakdown Testing TinySmallMediumLarge Camaleão250 SecurEngine250 243 Stash-It250 Total750 743 2,993 total images in the PNG stego category

14 IEEE-WVU, Anchorage - 2008  14 JPEG Data Set - Breakdown Training F5JPHideJStegOutguess Animals1,7322,127244436 Business3,779-12411 Maps3,361-11268 Natural5,2111,11323270 Tourist4,9681,721268160 Vacation2,96035310035 Total22,0115,3141,080780 29,185 total images in the JPEG stego category

15 IEEE-WVU, Anchorage - 2008  15 JPEG Data Set - Breakdown Training Animals-Non-modified61 Business-Non-modified31 Maps-Non-modified28 Natural-Non-modified58 Tourist-Non-modified67 Vacation-Non-modified25 Misc-Non-modified1,996 Misc-Append-modified665 Misc-Crop-modified666 Misc-Overlay-modified662 Total4,259 29,185 total images in the JPEG stego category

16 IEEE-WVU, Anchorage - 2008  16 JPEG Data Set - Breakdown Testing TinySmallMediumLarge F5250 JPHide240322318101 Jsteg198202199198 Outguess 0.2481421-- Outguess 0.13491425-- Total1,6601,620767549 4,596 total images in the JPEG stego category

17 IEEE-WVU, Anchorage - 2008  17 Sample Usage: stegdetect JPEG Training Set Detected, CDetected, INo Steg Clean-3603809 F52201100 JPHide4506604204 JSteg63842121 Outguess 0.1322010295 Outguess 0.2135237 Detected, C: correct algorithm detected Detected, I: incorrect algorithm detected Overall false detect rate for the clean image set is 8.6%

18 IEEE-WVU, Anchorage - 2008  18 Sample Usage: stegdetect JPEG Testing Set Detected, CDetected, INo Steg Clean-89979 F50216784 JPHide333153495 JSteg4443530 Outguess 0.1320631679 Outguess 0.23237833 Overall false detect rate for the clean image set is 8.0%

19 IEEE-WVU, Anchorage - 2008  19 Sample Usage: stegdetect Detailed results for JPHide Test Set LargeMediumSmallTiny Detected, C51249258 Detected, I47613510 Negative38262222

20 IEEE-WVU, Anchorage - 2008  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 IEEE-WVU, Anchorage - 2008  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 IEEE-WVU, Anchorage - 2008  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 IEEE-WVU, Anchorage - 2008  23 Download! http://www.liv.ic.unicamp.br/wvu/datasets.php


Download ppt "IEEE-WVU, Anchorage - 2008  1 The Unseen Challenge Data Sets Anderson Rocha Walter Scheirer Siome Goldenstein Terrance Boult."

Similar presentations


Ads by Google