Made it about 15 times by hand before I realized the depth this could go. Fun, but I assume I am missing an automated decoding tool?
Same. Not seeing anything pre-built, assuming something like PIL needed? The varying colors would seem to throw a lot of extra complexity into just recognizing shapes, no? So I too assume I'm not searching for the right keywords on Google here.
Any little hints/nudges. Recursive solution would seem to be needed, but getting the code from the images as they change size, color palette, etc seems very complex if not having done image processing programming before.
I went after PIL and pytesseract (can't get it to work) but that seems capable of OCR. From there you could loop and process morse, but certainly wouldn't rate that as easy
For anyone needing help, PIL is all you need to solve this. The images may all be different, but individually they are simple and if you create the right set of rules you should be able to read them all with minimal code.
For anyone needing help, PIL is all you need to solve this. The images may all be different, but individually they are simple and if you create the right set of rules you should be able to read them all with minimal code.
Thank you! Just seeing that you were able to do that with PIL is very helpful and does not spoil anything. :-)
Tried manually doing the first level to get a feel, but even that password doesn’t seem to work. I think I need a nudge.
GCIH | GCIA
If you need help with something, PM me how far you've got already, what you've tried etc (I won't respond to profile comments, or on box release night). And remember to +respect me if I helped you ; )
For those wondering, I had the same issue and it had to be in lowercase.
Had that issue too.
Finally finished this one. Not too difficult once I actually read the name of the challenge. Annoyingly, after the first 200 archives, it would fail to read the password every 100 or so and I'd have to restart it, not sure why.
GCIH | GCIA
If you need help with something, PM me how far you've got already, what you've tried etc (I won't respond to profile comments, or on box release night). And remember to +respect me if I helped you ; )
Is there anyone have trouble with file 497 ? My script doesn't give a correct password. I tried to extract the morse code manually, and i got sihhhi************siihssh, but still incorrect.
It was a fun coding exercise. Working out an algorithm to distinguish the Morse code pixel from the background pixel was especially rewarding after you realized the colors may all looked similar to the naked eye until you zoomed in to the pixel level.
Is there anyone have trouble with file 497 ? My script doesn't give a correct password. I tried to extract the morse code manually, and i got sihhhi************siihssh, but still incorrect.
The trick is you need to make your OCR function adaptable for any change it may happen in the image. Try to see what patterns appear and apply them in order to decode it correctly
For those not inclined toward perl or python solutions, you can make partial solution work using shell commands. It won't be pretty. But basically, consider that there are some old image file formats that lend themselves to much easier cut⦚ing and typical unix-y text processing.
For those wondering, I had the same issue and it had to be in lowercase.
By the way, great challenge ! Very nice to train your dev skills.
I won't call it easy challenge tho.
Thanks for this one!
Finding the best method to retrieve the password was the hardest part.
Tried to Google OCR solutions, but this was taking me way to far. Decided to go back to the basics and think about fg and bg colours.
Once I got the Python script ready to get the password, I used my solution for the Eternal Loop challenge to automate the unzipping.
easy, all it took were the 20 most cursed lines of python you'll never see 😁😁😁
i enjoy challenges like this, they're great practice for hacking together quick and dirty scripts for weird real world situations in which you encounter weirdly formatted data
I've been on this challenge for several hours already, I quickly understood the challenge or finally the first part how to extract all the zip, but the problem, doing the challenge manually, it's quickly a waste of time , I make a python script to recover the text with pytesseract and PIL , i try to decode the Morse text but it does not work, I do not know if the pixelation is too small, even after doing mogrify -set units PIxelPerInch -density 150 pwd.png for change the image size, but it still doesn't work, can someone help me.
thank you
Valiant, nothing is impossible. Lock by lock and one after the other is the key. You cannot open door number 9 until you have unlocked number 8.
Comments
Respect me first ;D
I somehow enjoyed this... :P
Made it about 15 times by hand before I realized the depth this could go. Fun, but I assume I am missing an automated decoding tool?
OSCP, GWAPT, GCIH, CISSP, Sec+ ||||| Starting Mar 20: AWAE
parityinfosec.com
Type your comment> @imag1ne said:
Same. Not seeing anything pre-built, assuming something like PIL needed? The varying colors would seem to throw a lot of extra complexity into just recognizing shapes, no? So I too assume I'm not searching for the right keywords on Google here.
Any little hints/nudges. Recursive solution would seem to be needed, but getting the code from the images as they change size, color palette, etc seems very complex if not having done image processing programming before.
I went after PIL and pytesseract (can't get it to work) but that seems capable of OCR. From there you could loop and process morse, but certainly wouldn't rate that as easy
OSCP, GWAPT, GCIH, CISSP, Sec+ ||||| Starting Mar 20: AWAE
parityinfosec.com
For anyone needing help, PIL is all you need to solve this. The images may all be different, but individually they are simple and if you create the right set of rules you should be able to read them all with minimal code.
Type your comment> @snuggles said:
Thank you! Just seeing that you were able to do that with PIL is very helpful and does not spoil anything. :-)
Great challenge to test your programming skills. Indeed, PIL is all you need to solve this, but you have to get a little bit creative.
Very cool challenge, I enjoyed the process, and found a cool but easy way to bypass the always changing values.
Thank you @swani really cool one
It looks daunting at first, but it is actually not too difficult.
The key is to simplify everything as much as you can before processing.
Honestly one of the most fun challenges I've done in a while.
PIL really is all you need. Just don't overthink what you're trying to do.
PM for help.
If you need help with something, PM me how far you've got already, what you've tried etc (I won't respond to profile comments, or on box release night). And remember to +respect me if I helped you ; )
Anyone else having trouble with file 986? PW in image does not match the archive if you do it the same way as for the other files..
....nvm, I was retarded. Thanks @HKLM
Type your comment> @Ljugtomten said:
For those wondering, I had the same issue and it had to be in lowercase.
By the way, great challenge ! Very nice to train your dev skills.
I won't call it easy challenge tho.
@Crafty said:
Had that issue too.
Finally finished this one. Not too difficult once I actually read the name of the challenge. Annoyingly, after the first 200 archives, it would fail to read the password every 100 or so and I'd have to restart it, not sure why.
If you need help with something, PM me how far you've got already, what you've tried etc (I won't respond to profile comments, or on box release night). And remember to +respect me if I helped you ; )
Is there anyone have trouble with file 497 ? My script doesn't give a correct password. I tried to extract the morse code manually, and i got sihhhi************siihssh, but still incorrect.
It was a fun coding exercise. Working out an algorithm to distinguish the Morse code pixel from the background pixel was especially rewarding after you realized the colors may all looked similar to the naked eye until you zoomed in to the pixel level.
Write-ups | Discord - limbernie#0386
Type your comment> @htw6666 said:
The trick is you need to make your OCR function adaptable for any change it may happen in the image. Try to see what patterns appear and apply them in order to decode it correctly
Was fun...
around 100 lines of python
@Crafty thanks for the hint
What a awesome challenge! I really enjoyed this one!
was pretty fun. I used perl instead of python. I couldn't get OCR to work so did it probably an over-complicated way but got it in the end.
For those not inclined toward perl or python solutions, you can make partial solution work using shell commands. It won't be pretty. But basically, consider that there are some old image file formats that lend themselves to much easier cut⦚ing and typical unix-y text processing.
OMG, I started the challenge right now
it is going to be fun
.
Type your comment> @Crafty said:
Thanks for this one!
Finding the best method to retrieve the password was the hardest part.
Tried to Google OCR solutions, but this was taking me way to far. Decided to go back to the basics and think about fg and bg colours.
Once I got the Python script ready to get the password, I used my solution for the Eternal Loop challenge to automate the unzipping.
easy, all it took were the 20 most cursed lines of python you'll never see 😁😁😁
i enjoy challenges like this, they're great practice for hacking together quick and dirty scripts for weird real world situations in which you encounter weirdly formatted data
Really nice challenge! I enjoyed in this one!
Good to learn different things.
Wrote script that works up until flag_986 where it fails, even manually.
Edit: seems like we are case sensitive.
I tried all function of PIL but can't get any information.
Pls give me a hint!
Edit: Morse and PIL, I got it!
@swani you had me wrote my worst python script ever. Fun though
thanks !
(pm for help, though quite easy)
Hello,
I've been on this challenge for several hours already, I quickly understood the challenge or finally the first part how to extract all the zip, but the problem, doing the challenge manually, it's quickly a waste of time , I make a python script to recover the text with pytesseract and PIL , i try to decode the Morse text but it does not work, I do not know if the pixelation is too small, even after doing
mogrify -set units PIxelPerInch -density 150 pwd.png
for change the image size, but it still doesn't work, can someone help me.thank you
Valiant, nothing is impossible.
Lock by lock and one after the other is the key. You cannot open door number 9 until you have unlocked number 8.