skotz / captcha-breaking-library Goto Github PK

There's currently a massive learning curve to get your first project going. Create a walk-through to help new users.

Hi Scott, during the training I see that:
Image name will be the value in captcha but if there is no other way to name the file we will not able to train multiple image with same value but different in format. Say I have image: ABC.bmp & if another image name ABC.bmp with different format, I will not able to train the solver with 2nd one if not overwrite 1st.

Do you see any way we can solve this?

Thank you.

Thank Skotz for your very good lib.

I'm working on an captcha which I can work on the filter to have clean final image but currently I don't see how to train the solver and get db?

Actually I have just read previous Issue ticket but not sure how to implement the training properly.

Such as we need to change script to Train & Save instead of Solve?
TRAIN, "I:\qqq"
SAVE, I:\qqq\new.db"

I do see we need a test pool of as big as possible images but how can tell Solver which is good which is bad?

On your script in previous Issue:
TRAIN, "qqq"
FULLTEST, "qqq", "test.txt"

I even see the result: TOTAL CORRECT: 213 (213/266 = 80.08%)

Is that we are manually calculate or the lib support that?

Thank for your support in advanced.

My script.
SetMode, all
SetupSegmenter, BLOB, 30, 40, 4
SetupSolver, BVS, "ABCDEFGHIJKLMNOPQRSTUVWXYZ", 50, 60

DefinePreconditions
Resize, 400, 140
RemoveSmallBlobs, 10, 4, 4
BlackAndWhite
Grow
Grow
Median
EndPreconditions

Solve, %IMAGE%

You currently need to continuously play around with image manipulation functions to correctly segment an image, saving intermediate steps to image files for examination. Create an integrated environment with a wizard-like tool for tweaking segmentation settings.

Can you explain how BVS work?

Please, write document "How to teach solver"

SETMODE, ALL
SETUPSEGMENTER,  BLOB, 3, 15, 6
SETUPSOLVER,     MNN, "0123456789", 20, 20
DEFINEPRECONDITIONS
    REMOVENONCOLOR, 60
    MEDIAN, 3
    BINARIZE, 80
    REMOVESMALLBLOBS, 180, 4, 14
ENDPRECONDITIONS
TRAIN, "I:\qqq"
SAVE, I:\qqq\new.db"

new.db file too small (less then one capthca file 60kb bmp)
Captcha images very good segmented.
FULLTEST show 0 from 266 images recognized

I really need such a tool to make my work easier.

I hope u keep going.

Thank you.

Original issue reported on code.google.com by [email protected] on 14 Nov 2012 at 7:04

Right now you basically need to create two CBL scripts to solve a CAPTCHA: one for training the solver and one for solving individual images. Create some way to merge these two ideas together to prevent confusion.

TESTSEGMENT always falling