Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/tmcclintock/imagesfrompdf
A python script to pull images from a PDF, and then make backgrounds transparent.
https://github.com/tmcclintock/imagesfrompdf
Last synced: 14 days ago
JSON representation
A python script to pull images from a PDF, and then make backgrounds transparent.
- Host: GitHub
- URL: https://github.com/tmcclintock/imagesfrompdf
- Owner: tmcclintock
- License: mit
- Created: 2020-07-05T17:20:29.000Z (over 4 years ago)
- Default Branch: master
- Last Pushed: 2020-07-10T23:04:57.000Z (over 4 years ago)
- Last Synced: 2024-12-14T17:09:32.858Z (about 1 month ago)
- Language: Python
- Size: 78.1 KB
- Stars: 3
- Watchers: 2
- Forks: 1
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# ImagesFromPDF
A python script to pull images from a PDF, and then make
backgrounds transparent with `ImageMagick`. Images are saved
as `png` files.
## Requirements
The first requirement is [`pdfreader`](https://github.com/maxpmaxp/pdfreader),
which is a Python package for manipulating PDFs. It can be installed with
`pip` using
```bash
$ pip install pdfreader
```
There are other install options detailed in that repository. Note that
you must be using Python 3.X to use `pdfreader`.
The other requirement is [`ImageMagick`](https://imagemagick.org/index.php),
which is a command-line program for editing images. It is used in
this script to remove white and/or black backgrounds from images, but
is not required to simply extract images.
`ImageMagick` must be installed following the instructions in their
documentation. You can use `brew` if you have a Mac, but the
steps are slightly more complicated for Windows or Unix users.
## Usage
The script in thie repository is `extract_images_from_pdf.py` and it
can be ran from the command line with
```bash
$ python extract_images_from_pdf.py /path/to/MyFile.pdf
```
You can run this script with a number of flags that you can set
to enable different features. These include:
* `-o` or `--output`; controls the output directory, by default it will output to `_images/` where the input file is called `/path/to/.pdf`.
* `-v` or `--verbose`; set to `True` by default, controls the amount of output provided.
* `-fp` or `--first_page`; default 0, first page to export from.
* `-lp` or `--last_page`; default 1000, last page to export from.
* `-mw` or `--min_width`; default 200, minimum pixel width of pictures to export.
* `-mh` or `--min_height`; default 200, minimum pixel height of pictures to export.
* `-xw` or `--max_width`; default 1210, maximum pixel width of pictures to export.
* `-xh` or `--max_height`; default 1517, maximum pixel height of pictures to export.
* `-mt` or `--make_transparent`; default `False`, flag to attempt to make backgrounds transparent.
* `-wt` or `--white_to_trans`; default `True`, if `-mt=True` then set this flag to make white pixels transparent
* `-bt` or `--black_to_trans`; default `True`, if `-mt=True` then set this flag to make black pixels transparent.
* `-wf` or `--white_fuzz`; default 1, if white pixels are made transparent, sets the `ImageMagick` fuzz percentage (i.e. sets _almost_ white pixels to transparent as well). Can be 0-100.
* `-bf` or `--black_fuzz`; default 1, if black pixels are made transparent, sets the `ImageMagick` fuzz percentage (i.e. sets _almost_ black pixels to transparent as well). Can be 0-100.
* `-ims` or `--image_string`; default is `"Im"`, string that appears in all image names used to indicate which images to pull from the document.
Note that the max pixel height and width values correspond to just under the size of pages in standard [Pathfinder bestiaries](https://paizo.com/products/btpy8auu?Pathfinder-Roleplaying-Game-Bestiary).
## Example usage
I have a PDF of the [Pathfinder Bestiary](https://paizo.com/products/btpy8auu?Pathfinder-Roleplaying-Game-Bestiary) released by Paizo. I was able to use this script to pull out all monster images into `png` files with transparent backgrounds using:
```bash
$ python extract_images_from_pdf.py Bestiary1.pdf -mt=True
```
These images can then be inserted into virtual tabletop software.
## Tests
This sript has been tested on Python 3.7 on a Mac.
## Contributing
PRs are welcome. Big upgrades could include giving sensible names to images based on nearby text in the documents, as well as a test suite that I was too lazy to make.