captcha-vision-engine 0.1.1

OCR-based solvers for noisy and grid-style image CAPTCHAs

captcha-solver

OCR-based Python library to solve two common image CAPTCHA styles. Ready for PyPI.

Requires Tesseract OCR installed on your system (not only the pytesseract pip package).

---

CAPTCHA types this library solves

1. Image CAPTCHA — solveImageCaptcha

Clear character row (e.g. 6 characters).

2. Noisy CAPTCHA — solveImageNoiseCaptcha

Heavy background noise, diagonal lines, and speckles. Use when the text is obscured but still readable. Uses segmentation + per-character OCR. Returns one result or 25 candidate results to try until the form accepts one.

---

Install

pip install captcha-solver

System dependency: Install Tesseract OCR.

Platform	Command
Ubuntu/Debian	sudo apt-get install tesseract-ocr
macOS	brew install tesseract
Windows	Tesseract at UB Mannheim — then set TESSERACT_CMD to the executable path if needed.

---

Usage

Class-based API

from captcha_solver import CaptchaSolver

solver = CaptchaSolver()

# Image CAPTCHA → one result
text = solver.solveImageCaptcha("path/to/Image.png")

# Noise CAPTCHA → one result (optional: num_chars, uppercase)
text = solver.solveImageNoiseCaptcha("path/to/noisy_captcha.png")

# Grid CAPTCHA → 25 candidate results (try each until form accepts)
results = solver.solve_grid_five_results("path/to/grid_captcha.png")
for i, r in enumerate(results, 1):
    print(f"{i}: {r}")

Function-based API

from captcha_solver import solveImageCaptcha, solveImageNoiseCaptcha

text = solveImageCaptcha("path/to/noisy_captcha.png")
text = solveImageNoiseCaptcha("path/to/grid_captcha.png")

# With optional params (grid only)
text = solveImageNoiseCaptcha("path/to/grid_captcha.png", num_chars=6, uppercase=False)

Optional: Tesseract path (e.g. Windows)

import os
os.environ["TESSERACT_CMD"] = r"C:\Program Files\Tesseract-OCR\tesseract.exe"

from captcha_solver import CaptchaSolver
solver = CaptchaSolver()
# ... use as above

---

Grid solver options

Parameter	Type	Default	Description
num_chars	int | None	5	Expected number of characters; result is trimmed/padded to this length.
uppercase	bool	True	If True, return uppercase; if False, return lowercase.

---

Publish to PyPI

# Install build and twine
pip install build twine

# Build package
python -m build

# Upload to Test PyPI (optional)
twine upload --repository testpypi dist/*

# Upload to PyPI
twine upload dist/*

---

Development

git clone https://github.com/yourusername/captcha-solver
cd captcha-solver
pip install -e ".[dev]"

---

License

MIT.