Breaking captchas with imageMagick and gocr

cool5now · Feb 11, 2010

Hi,

I am trying to write a module that can crack a fairly simple captcha. The one I am focusing on to start with is http://captchas.net/ a grainy monochrome captcha where the individual letters are slightly rotated.

I've found a way to download hundreds of different images of the same captcha text. i.e. different images but same text.

The plan is to use imagemagick to clean up the image and then train gocr to convert it to text. Do this 500 times check which is the most popular answer gocr has given me and then send it back to the registration script.

The problem I'm having is cleaning the image. The best I can do at present is:
system 'convert ' + $capfile + ' -paint 0.01 -paint 0.01 -paint 0.01 -paint 0.01 -paint 0.01 -paint 0.01 -quality 100 -shave 40x20 ' + $capfile

This removes the background blocks up the text and trims the borders. But it isn't perfect by any means and when I start to train gocr even I can't guess what letters like e and a are because they've had the middles filled in. I've read about the threshold command but can't seem to get it to work.

Is there any other image magick commands that could help clean up this file?

I've read the post at bluehatseo and some other ones on the net but still can't find a way to really clean up the image. If some one could give me a little shove in the right direction I'd be very grateful.

cool5now · Feb 11, 2010

Getting better:
convert blog420.png -colorspace Gray -gaussian-blur 9.0 -level 0,25%,0,1 -depth 1 -paint 0.01 test.png

minute80 · Feb 11, 2010

This guy has great tutorials about the same subject:

Code:

http://www.bluehatseo.com/index.php?s=captcha&submit=Search

cool5now · Feb 11, 2010

Yeah but programming in C makes my head spin even on simple stuff like this. Thanks for your help though. I got it sorted in the end. If anyone else is interested this is the command I used: system 'convert ' + $capfile + ' -colorspace Gray -gaussian-blur 9.0 -level 0,30%,0,1 -paint 1 -depth 1 -quality 100 -shave 40x20 ' + $capfile

Trained gocr and have got it up and running. I've tested the program around a hundred times... so far 100% success rate, happy days.

dudeshane01 · Mar 10, 2011

Hi Buddy
Can this captcha breaker be used with any seo tools too?
I am really interested to know more about it if it does.

Autumn · Mar 10, 2011

Nice work!

I used to do what you're doing back in the day but these days I just go with captcha breaking services. At the end of the day if you value your time, I reckon it's cheaper to just pay for captcha breaking and then you can crank through the http part of whatever you want to automate in no time at all. It's a lot of work just to break one captcha and captcha services are super cheap.

Here are a few my old imagemagick strings, you might see an option you like:

Code:

convert penisbot.png -fill white -opaque '#6ecefe' -opaque '#fede6e' -opaque '#72fe6e' penisbot2.gif


convert only4porn.jpg -shave 4x4+0+0 -trim +repage -fuzz 3000 -fill black -opaque '#000000' -black-threshold 60000 only4porn2.jpg


convert c.gif -crop 26x26+92+2 -trim +repage -fuzz 3000 -fill white -opaque '#cdfe00' -black-threshold 60000 4.gif

Autumn · Mar 10, 2011

I assume you're using PHP, so you might get some value out of this too. This code picks the colours from the specified pixels of an image, if you're lucky enough to find a captcha where they try to obfuscate the image with silly colours that don't appear in the text itself. Then you can replace those colours with imagemagick.

PHP:

// Original image
$image = 'image.gif';

// Get background colours from image
// x and y coordinates to pick a pixel from, for each quarter of the image
$x_vals = array(5, 35, 65, 95);
$y = 4;

// Create GD image handler
$my = ImageCreateFromGif($image);

// Retrieve pixel colour of each point
foreach($x_vals as $x) {
	$rgb = ImageColorAt($my,$x,$y);

	// Transform integer value to rgb
	$trans =  imagecolorsforindex($my, $rgb);
	//echo "{$trans['red']}/{$trans['green']}/{$trans['blue']}\n";

	// Convert rgb values to hex
	// Need to add a zero to the end of each value if strlen==1
	$red = dechex($trans['red']);
	if(strlen($red) == 1) { 
		$red = $red .'0';
	}
	$green = dechex($trans['green']);
	if(strlen($green) == 1) { 
		$green = $green .'0';
	}
	$blue = dechex($trans['blue']);
	if(strlen($blue) == 1) { 
		$blue = $blue .'0';
	}

	$bg_cols[] = "#$red$green$blue\n";
}

// Destroy image handler
imagedestroy($my);

Good luck!

Edit: damn, just saw the original date of the post. Oh well, hopefully this is of value to someone.

Breaking captchas with imageMagick and gocr

cool5now

Registered Member

cool5now

Registered Member

minute80

Regular Member

cool5now

Registered Member

dudeshane01

Newbie

Autumn

Elite Member

Autumn

Elite Member

Main Menu

Marketplace

Making Money

BlackHat World