A Python script that deletes redundant visually-similar images.
For a list of visually similar images it finds, it first deletes all but the largest by file-size. If still multile files remain, it deletes all but the oldest by modified-on.
These are the accepted options:
| Option | Description |
|---|---|
| -p | Mandatory. Path to the folder under which to recusively scan. |
| -r | The script creates a 'dups.txt' file as an intermediate while processing. This file holds the list of visually-similar filenames. Giving -r prevents deleting this intermediate file after processing. |
| -d | Dry run. Prints out the names of the files which would be deleted. Doesn't actually delete them. |
- findimagedupes
- Supports jpg, png and gif files
- The script can be trivially modified to support even more file types.
duplicate; image; remove; detect; find;