PyQArt - QArt Python implementation
introduction
QArt is a method of combining QrCode of an URL with any image, which was submitted in an article writen by Russ Cox on his personal website.
An Example(come from the article):
This repo is Python implementation of it.
Install
pip install pyqart
Note: Support Python3 only, please make sure you are using pip of Python3.
Usage
For code reuse, I split the lib to two part. One for generate normal QrCode, another for generate QArt.
The Qr Part
Use pyqr
CLI to create normal QrCode.
pyqr -p 5 -c 102 204 255 "Hello World!" -o qr.png
The options:
-p
for point size of QrQCode, by pixel, default is 3 pixel.-c
for color of point, default is black. Background color can be set with-g
option, default is white.
If you want show it in terminal, just don't provide -o
option:
pyqr "Hello World"
Then you will see:
The actual result you will see depends on your font setting, I'm using Dejavu Sans Mono.
Yes, it is only useful for small QrCode.
Run pyqr -h
for more options and their effect.
The Art Part
Use pyqart
CLI to create QArt. It may take a long time, please be patient :)
This is an example that mix my blog url and my Github avatar:
pyqart -v 8 -c 102 204 255 "http://0v0.link/" photo.jpg -o qart.png
My Github avatar:
The QArt Code:
Not meet your expectations? Try -n
option to pick point at random(default is pick low-contrast region pixels first):
pyqart -n -c 102 204 255 -v 8 "http://0v0.link/" photo.jpg -o qart-n.png
Still not satisfied? Use -y
option to enhance the accuracy of the central region by giving up the control of the edge pixels:
-y
and -n
can be used at the same time, but no obvious improvement.
Note: because that -y
option will only use data block, ignore error correction block,it reduce many many many calculate. It has about 30x to 100x speed up compare with no -y
option case. So I strongly recommend using -y
option whenever you needn't make a full picture fitting.
Use -r
option to set rotation degree, The controllable data region can be changed into a horizontal area, it will make it easier to process very wide picture.
Run pyqart -h
for more options and their effect.
Use it in your codes as a module
Documentation is in preparation.
Gallery
python.org(used -d option, means dithering, see help message for more info.)
github.com
bilibili.com (An ACG videos website)
Halftone and HalfArt support
Halftone support added in version 0.1.0, and I made another new method which combined Halftone and QArt, so I call it HalfArt temporarily.
Arguments for all methods
The following code shows arguments to get output image of all kind of method:
from pyqart import QArtist, QrHalftonePrinter, QrImagePrinter, QrPainter
QR_VERSION = 10
POINT_PIXEL = 3
artist = QArtist('http://www.nankai.edu.cn/', 'example.jpg', QR_VERSION)
painter = QrPainter('http://www.nankai.edu.cn/', QR_VERSION)
artist_data_only = QArtist('http://www.nankai.edu.cn/', 'example.jpg',
QR_VERSION, only_data=True)
# normal
QrImagePrinter.print(painter, path='normal.png', point_width=POINT_PIXEL)
# Halftone
QrHalftonePrinter.print(painter, path='halftone.png', img='example.jpg',
point_width=POINT_PIXEL, colorful=False)
# Halftone colorful
QrHalftonePrinter.print(painter, path='halftone-color.png', img='example.jpg',
point_width=POINT_PIXEL)
# Halftone pixel
QrHalftonePrinter.print(painter, path='halftone-pixel.png', img='example.jpg',
point_width=POINT_PIXEL, colorful=False,
pixelization=True)
# QArt
QrImagePrinter.print(artist, path='qart.png', point_width=POINT_PIXEL)
# QArt data only
QrImagePrinter.print(artist_data_only, path='qart-data-only.png',
point_width=POINT_PIXEL)
# HalfArt
QrHalftonePrinter.print(artist, path='halfart.png', point_width=POINT_PIXEL)
# HalfArt data only
QrHalftonePrinter.print(artist_data_only, path='halfart-data-only.png',
point_width=POINT_PIXEL)
Result example for all method
Halftone | Halftone colorful | Halftone pixel |
QArt | QArt data only | |
HalfArt | HalfArt data only |
TODO
- Make QrPainter decided argument by itself.
- Art part
- CLI
- Package
- Halftone support
- self-made HalfArt method
- GUI
- Use Cython to accelerate Reed-Solomon error correction
- Docs
- Tests
Other Implementation
- Golang: qr by Russ Cox
- Java: qart4j by dieforfree
Acknowledgements
- All credit goes to Russ Cos, Thanks for his article and implement.
- Thanks for qart4j project by dieforfree,which helps me so much on how to implement the art part.
- Thanks to a series of articles named QR Code Tutorial in thonky.com, It's very detailed. Whenever I faced problem about encoding or error correction, I will go to it for help.
- Thanks to the Python programing language。
LICENSE
MIT.
See LICENSE.