The ddddocr library in Python that recognizes image/slider captchas with very high accuracy details

preamble

There are many types of CAPTCHA, which is commonly used as a means of anti-climbing, including: image CAPTCHA, slider CAPTCHA, and some other common CAPTCHA scenarios.

Recognize the CAPTCHA python library has a lot to use and is not simple, here is a simple and practical to recognize the CAPTCHA library ddddocr (with a younger brother ocr) library.

environmental preparation

The python version required is less than or equal to python version 3.9.

pip install

pip install ddddocr

Download the installation package is relatively large, generally use the domestic download source can speed up the download speed

pip install ddddocr -i /simple

github address/sml2h3/ddddocr

Quick Start

First, find a random English captcha, keep it to

code example

import ddddocr                       # Import ddddocr
ocr = ()              # Instantiation
with open('', 'rb') as f:     # Open the picture
    img_bytes = ()             # Read the picture
res = (img_bytes)  # Identify
print(res)

running result

has been able to recognize xnen, but there will be "welcome to use ddddocr, this project focuses on driving the industry volume ..." prompt, can add a parametershow_ad=False

import ddddocr                       # Import ddddocr
ocr = (show_ad=False)              # Instantiation
with open('', 'rb') as f:     # Open the picture
    img_bytes = ()             # Read the picture
res = (img_bytes)  # Identify
print(res)

CAPTCHA

Recognize the three types of CAPTCHA

code example

import ddddocr                       # Import ddddocr
ocr = (show_ad=False)              # Instantiation
with open('', 'rb') as f:     # Open the picture
    img_bytes = ()             # Read the picture
res2 = (img_bytes)  # Identify

print(res2) 
with open('', 'rb') as f:     # Open the picture
    img_bytes = ()             # Read the picture
res3 = (img_bytes)  # Identify
print(res3)

with open('', 'rb') as f:     # Open the picture
    img_bytes = ()             # Read the picture
res4 = (img_bytes)  # Identify
print(res4)

running result

giv6j
zppk
4Tskh

Slider CAPTCHA

The slider CAPTCHA scenario is the following scenario example

Start by keying out 2 images, namely and

Problem solving focuses on calculating the location of the gap

import ddddocr

det = (det=False, ocr=False, show_ad=False)

with open('', 'rb') as f:
    target_bytes = ()

with open('', 'rb') as f:
    background_bytes = ()

res = det.slide_match(target_bytes, background_bytes, simple_target=True)
print(res)

running result

{'target_y': 0, 'target': [184, 58, 246, 120]}

The four values of target are the upper-left and lower-right corners of the notch position to the left.

Recognize Chinese

Recognize text on pictures

import ddddocr
import cv2

det = (det=True)

with open("", 'rb') as f:
    image = ()

poses = (image)

im = ("")

for box in poses:
    x1, y1, x2, y2 = box
    im = (im, (x1, y1), (x2, y2), color=(0, 0, 255), thickness=2)

("", im)

Saved images

To this point this article on Python to recognize the picture/slider CAPTCHA very high accuracy ddddocr library details of the article is introduced to this, more related Python ddddocr library content, please search for my previous articles or continue to browse the following related articles I hope you will support me in the future!