Encode/decode images using Base64 or shuffle/recover the pixels of images.
This repository is a renewed implementation of Python code I saw a long time ago on CoolApk:
import matplotlib.pyplot as plt
import numpy as np
from PIL import Image
# Get the three-dimensional pixel channel matrix of the image
img = np.array(Image.open("C:/Users/user/Downloads/test.png"))
# First dimension
row_len = img.shape[0]
# Shuffle the dimension indices
row_index = np.random.permutation(row_len)
# Generate the chaotic image
img_chaos = img[row_index, :, :]
# Use sorting to unshuffle the image
img_sort = img[np.sort(row_index), :, :]
# Plot the chaotic and unshuffled images
plt.figure('Chaotic and Unshuffled Images')
plt.subplot(121)
plt.imshow(img_chaos)
plt.subplot(122)
plt.imshow(img_sort)
plt.show()And it can also be seen as an implementation of similar functions of the Android application 图片混淆 in Python.
-
The original image:
-
The shuffled image:
The
.npzfile key.npz stores the original positions of pixels. -
The recovered image:
The binaries are created with Nuitka:
# Show Nuitka version
python -m nuitka --version
# Package it on Linux
python -m nuitka --onefile --remove-output pixel_puzzle.py
# Package it on Windows
python -m nuitka --onefile --remove-output --windows-icon-from-ico="python.ico" pixel_puzzle.pyDeploy web applications using:
Here are some links to other similar projects that I am aware of:
Warning
Contents of this repository may cause some copyright issues. Please raise issues to let me know if that happens.
PixelPuzzle is a free, open-source software package (distributed under the GPLv3 license). Some of the references are as follows:
The Python Software Foundation owns the copyright of Python icon. The sample image used is downloaded from satchely doki doki literature club! natsuki.