Skip to content

Latest commit

 

History

History
36 lines (26 loc) · 1.99 KB

README.md

File metadata and controls

36 lines (26 loc) · 1.99 KB
Raw

Usage

The script can be run from the command line with the following arguments:

  • -i, --image: The path to the large image file. Example: "/home/user/Pictures/image.png"
  • -id, --input-dir: The path to the directory containing image files. Example: "/home/user/Pictures/Images"
  • -d, --dest: The new folder to save the cropped images. Example: "/home/user/Pictures/CroppedImages"
  • --color: The colors to check for in hexadecimal or RGB format. Default is black (#000000). Example: "#FFFFFF" "#FF0000" or "255,255,255" "255,0,0"
  • --color_threshold: The color threshold. If a pixel's color is above this threshold, it is considered as a color pixel. Default is 70. Example: 80
  • --color_percentage: The percentage of color pixels in a row for an image to be considered having too much of that color. Default is 10. Example: 20
  • --color_specs: The color specifications in the format "color,threshold,percentage". Example: "#FFFFFF,80,20" "#FF0000,70,10"
  • -bw, --black_white: Check for both black and white colors.
  • -b, --black: Check for black color. This is the default behavior.
  • -w, --white: Check for white color.
  • --height: The desired height of the cropped images. Default is 512. Example: 768
  • --width: The desired width of the cropped images. Default is 512. Example: 720

Example

Here is an example of how to run the script:

Quick Run:

python change_image512.py -id "./Images" -d "./output/"

This will check do the default behavior which is black threshold of 60, black percentage of 10. No white check, no color check.

Advanced:

python change_image512.py -id "/home/user/Pictures/Images" -d "/home/user/Pictures/CroppedImages" --color "#FFFFFF" "#FF0000" --color_threshold 80 --color_percentage 20 --color_specs "#FFFFFF,80,20" "#FF0000,70,10" -bw --height 768 --width 720

This will crop the images to 768x720, check for both black and white colors, and check for the colors "#FFFFFF" and "#FF0000" with the specified thresholds and percentages. ```