Image compression using Pillowscript

Prerequisites

Programmatic compression of images using existing algorithmscan be undertaken using a command-line script.

1. Download the lossy_compression.py script to your device
2. Move the script to a stable location, such as your userhome directory
3. Make the script executable chmod+x .../lossy_compression.py

• Install ImageMagick brew install imagemagick
• Install Pillow python3 -m pip install pillow
• Install Numba python3 -m pip install numba
• Install NumPy python3 -m pip install numpy
• Install openpyxl python3 -m pip install openpyxl

1. Bulk resize

To bulkresize all images in a directory, use Mac’s built-in sips tool. Navigate to the destination in Terminal and then run the followingscript:

sips -Z 1200 \*.jpg

*NB this script may inadvertently upscale images smallerthan the dimensions specified.

To avoid this, navigate to a top-level directory usingTerminal and use the following to search for all files that exceed thespecified dimension recursively, and then only apply the transformation tothose:

mogrify -resize '2000>' *.jpg 

2. Bulk set DPI

To bulk change the DPI of all images in a directory,navigate to the destination in Terminal and then run the following script:

find . -name "*.jpg" -exec mogrify -units "PixelsPerInch" -density 72\> {} \;

3. Dithering

Output formats:

• JPEG
• PNG
• WebP

Dithering modes:

• one_bit
• bayer_2x2
• cluster_4x4
• bayer_8x8
• floyd_steinberg_pil
• atkinson
• stucki
• jarvis_judice_ninke
• grayscale
• floyd_stinberg_dev

python3 "SCRIPT LOCATION" "IMAGE DIRECTORY" "lossy_compression" "DITHERING MODE" "FORMAT"

The script will create a new directory at the same level asthe source in which to store the derivatives generated. Additionally, it willproduce a spreadsheet listing the files and their relative transformation sizes.