Using ddddocr SDK for Captcha Recognition in Python

Introduction

When crawling websites that require login, you often encounter captchas such as simple letter codes, slider puzzles, or click‑based image selections. These are long‑standing challenges for web scrapers. The ddddocr SDK ("ddddocr brother OCR") is a free, open‑source solution that can handle all of them.

Installation

The SDK automatically installs the appropriate version for your environment. It requires Python 3.9 or lower.

<code>pip install ddddocr</code>

Usage

ddddocr can recognize three types of captchas. The following sections demonstrate each type.

Slider Captcha

The example uses Douban's slider captcha. The slider piece is stored in hycdn.png and the background with the slot in background.jpg .

<code>import ddddocr

det = ddddocr.DdddOcr(det=False, ocr=False)

with open('hycdn.png', 'rb') as f:
    target_bytes = f.read()
with open('background.jpg', 'rb') as f:
    background_bytes = f.read()

res = det.slide_match(target_bytes, background_bytes, simple_target=True)
print(res)
</code>

Sample output:

<code>{'target_y': 0, 'target': [486, 126, 622, 262]}</code>

Click‑Based Captcha

This example uses a click‑based captcha from NetEase login.

<code>det = ddddocr.DdddOcr(det=True)

with open('eb.jpg', 'rb') as f:
    image = f.read()

poses = det.detection(image)
im = cv2.imread('eb.jpg')
for box in poses:
    x1, y1, x2, y2 = box
    im = cv2.rectangle(im, (x1, y1), (x2, y2), color=(0, 0, 255), thickness=2)
cv2.imwrite('result.jpg', im)
</code>

The resulting image shows the detected click positions.

Alphanumeric Captcha

The image is taken from a Google search.

<code>ocr = ddddocr.DdddOcr(old=True)

with open('z1.jpg', 'rb') as f:
    image = f.read()

res = ocr.classification(image)
print(res)
</code>

Sample output:

<code>3n3d
8342
</code>

Conclusion

ddddocr makes solving captchas simple and accessible, allowing developers who are not familiar with OpenCV, PyTorch, or TensorFlow to quickly bypass login verification challenges. Users are encouraged to share other effective OCR tools in the comments.