Loading
Cookies help us deliver our services. By using our services, you agree to our use of cookies. Learn more

crowdAI is shutting down - please read our blog post for more information

ImageCLEF 2019 Security - Forged file discovery

Identify forged images


Completed
8
Submissions
59
Participants
1727
Views

Potential problem with training set for Task 1..?

Posted by ABigPancake 10 months ago

Hi

I downloaded the training set for Task 1 last night it seems that the .pdf containing the images are not really pretending to be .pdf files. If you use a .pdf reading tool it is clear that a file contains image data in some format. Is it supposed to be like this or did something go wrong with the dataset creation?

Examples according to ground truth file for Task 1:

A .pdf containing image data:

pdfinfo 1200_01.pdf Syntax Warning: May not be a PDF file (continuing anyway) Syntax Error: Couldn’t find trailer dictionary Syntax Error: Couldn’t find trailer dictionary Syntax Error: Couldn’t read xref table

A .pdf containing pdf data.

pdfinfo 1201_01.pdf Title:
Author: hetzaki Creator: Acrobat PDFMaker 5.0 για το Word Producer: Acrobat Distiller 5.0.5 (Windows) CreationDate: Mon May 12 07:57:15 2003 BST ModDate: Mon May 12 06:58:27 2003 BST Tagged: yes UserProperties: no Suspects: no Form: none JavaScript: no Pages: 145 Encrypted: no Page size: 595 x 842 pts (A4) Page rot: 0 File size: 1714218 bytes Optimized: no PDF version: 1.3

Posted by Konstantinos Karampidis  9 months ago |  Quote

Hello. Everything is fine with the dataset . The purpose of the task is to identify the correct file type also in a digital forensic approach and not only the altered images. The latter could be very easy. The first is much more difficult. For example the most used forensic software like Autopsy, Encase etc. fail to identify correctly a forged file . In any case we apologize for the late answer

Posted by abcrowdai  8 months ago |  Quote

What do we have to submit ? Only a text file or our python code ? Please reply as soon as possible.

Posted by Konstantinos Karampidis  7 months ago |  Quote

What do we have to submit ? Only a text file or our python code ? Please reply as soon as possible. You have to submit a text file until 15/5/19. You can see how it must be structured in our web site https://www.imageclef.org/2019/security and in the description of each task also. In there you can also see the dates for the submission of abstracts etc.

Posted by Konstantinos Karampidis  7 months ago |  Quote

@abcrowdai > > What do we have to submit ? Only a text file or our python code ? Please reply as soon as possible. You have to submit a text file until 15/5/19. You can see how it must be structured in our web site https://www.imageclef.org/2019/security and in the description of each task also. In there you can also see the dates for the submission of abstracts etc.