A simple CLI program for converting images into text. While ASCII is in the name, it works with any unicode characters, and the specific characters used can easily be customised using the chars.json file.
usage: img_to_ascii.py [-h] [-o OUT_PATH] [-w WIDTH] [-c CHARS_IDX] [-i] [-p PIXEL_WIDTH] [-s] image_path chars_path
Convert an image into ASCII (or unicode) text
positional arguments:
image_path Path to the image to convert
chars_path Path to the chars file
options:
-h, --help show this help message and exit
-o OUT_PATH, --out_path OUT_PATH
File to output to. If left blank, will output to console.
-w WIDTH, --width WIDTH
Width to scale the image to. Default is 300px
-c CHARS_IDX, --chars_idx CHARS_IDX
Index of the 'pallet' in the given chars json file
-i, --invert If included, image will be inverted
-p PIXEL_WIDTH, --pixel_width PIXEL_WIDTH
How many characters to place per pixel. Should be adjusted based on font aspect ratio.
-s, --silent If included, only the output will be printed to the console
The following command will convert example/ubuntu.png into text, using the pallet at index 2 of chars.json. The image will be scaled to a width of 125px, and there will be 2 characters per pixel. The output will be saved to examples/ubuntu_out.txt.
$ python image_to_ascii.py examples/ubuntu.png chars.json -o examples/ubuntu_out.txt -c 2 -w 125 -p 2
Colour "pallets" are defined in the chars file. An example chars file is included in chars.json. chars files should be json files that contain a single array of strings. Each character in the string will be used as a shade of gray to create the image, meaning the darkest pixels will be represented with the first character, and the brightest pixels will be represented with the last character.
The images used in examples/ are not my own. Attribution for these images is included below, as needed.
Attribution: [email protected] Larry Ewing and The GIMP