How to Crack Image CAPTCHAs with Python: From PIL to pytesser

1. Discussion

Recognizing graphical CAPTCHAs is essential for web crawling and involves computer graphics, machine learning, computer vision, and artificial intelligence. Computer graphics studies how to represent, compute, process, and display geometric elements (points, lines, surfaces) and attributes such as color.

For CAPTCHA cracking, key knowledge includes handling 2‑D elements and color‑difference analysis. Common tools: Support Vector Machine (SVM), OpenCV, image editors (Photoshop, GIMP), and the Python Imaging Library (PIL).

2. Installing PIL

On Debian/Ubuntu Linux you can install directly with: sudo apt-get install python-imaging On other Linux distributions use sudo easy_install PIL or pip install Pillow. Windows users can download the installer from the official site; for 64‑bit systems use the Pillow wheel from the unofficial binaries page.

3. General Workflow

The typical steps for CAPTCHA recognition are:

Image denoising

Image segmentation

Text extraction

3.1 Denoising

Convert the image to grayscale (e.g., using the I component of HSV) and then binarize it. Example code:

from PIL import Image
im = Image.open('7039.jpg')
imgry = im.convert('L')
imgry.show()
threshold = 140
table = []
for i in range(256):
    if i < threshold:
        table.append(0)
    else:
        table.append(1)
out = imgry.point(table, '1')
out.show()

The original CAPTCHA image:

Grayscale conversion result:

Binary image after thresholding:

3.2 Segmentation

Segmentation is the most challenging part; for simple CAPTCHAs it may be skipped, while complex ones (e.g., Google reCAPTCHA) often require character splitting, which is beyond the scope of this tutorial.

4. Using pytesser for OCR

pytesser is a Python wrapper for Google’s Tesseract OCR engine. Install PIL, download pytesser and the Tesseract OCR engine, replace the tessdata folder, and adjust the import statement to from PIL import Image. Also create an empty __init__.py file in the pytesser directory.

Example usage:

from PIL import Image
from pytesser import pytesser
image = Image.open('7039.jpg')
print pytesser.image_file_to_string('7039.jpg')
print pytesser.image_to_string(image)

pytesser also supports other languages such as Chinese.