A minimal working project showing the problem would be appreciated to help in debugging.

The project is composed of 13 tex files and 7-10 images. There is a source.tex file that I am including in the attachment and other sub-files are added using /include command.

I am syncing the whole folder on Dropbox now and plan to post a public share link when it's fully synchronized, some 30min from time of posting.

Thank you for taking the time to assist with this.

The project is now fully synchronized, here is the link:

https://www.dropbox.com/sh/q434026ogc3184b/zUrm4yOYpw

Thanks. Please try to remove the DOC-corrected.pdf file from the project directory (or rename it or move it away if you think you will need it some day), and then re-run annotating process (tick off and on the compose option box if necessary to force re-run). It seems that the DOC-corrected.pdf file is not the right one (I will try to see when it has been generated like this).

Thank you, it worked right away!

I think I understand what happened. I reconstructed my steps and realized this was all due to a
simple silly mistake I made.

At one point I have updated the exam source file to include the funny picture at the end.
Then I thought I saved the new version to the USB, but apparently didn't.
Then I printed the exam from a USB, but apparently it was the wrong version! The printed version was the
older iteration, the one without the picture at the end with everything else being exactly the same.

This change hasn't influenced the order of the questions and the layout of the worksheet, so all the scans
were recognized and graded properly. That is why I didn't even notice the mishap.
However, the extra image disturbed the annotation process.

I am very sorry to waste your valuable time with a simple user mistake like this.

Regards,

Niko Z.