# page-dewarp
Document image dewarping library using a cubic sheet model
Python 3 library for page dewarping and thresholding,
[available on PyPI](https://pypi.org/project/page_dewarp/).
## Requirements
Python 3 and NumPy, SciPy, SymPy, Matplotlib, OpenCV, and TOML Kit are required to run `page-dewarp`.
- See [`CONDA_SETUP.md`](https://github.com/lmmx/page-dewarp/blob/master/CONDA_SETUP.md) for
an example conda environment
- If you prefer to install everything from pip, run `pip install page-dewarp`
This library was renovated from the [original (2016) Python 2 script](https://github.com/mzucker/page_dewarp/)
by Matt Zucker, which you can use if you are somehow still using Python 2.
## Usage
```
usage: page-dewarp [-h] [-d {0,1,2,3}] [-o {file,screen,both}] [-p]
[-vw SCREEN_MAX_W] [-vh SCREEN_MAX_H] [-x PAGE_MARGIN_X]
[-y PAGE_MARGIN_Y] [-tw TEXT_MIN_WIDTH]
[-th TEXT_MIN_HEIGHT] [-ta TEXT_MIN_ASPECT]
[-tk TEXT_MAX_THICKNESS] [-wz ADAPTIVE_WINSZ]
[-ri RVEC_IDX] [-ti TVEC_IDX] [-ci CUBIC_IDX]
[-sw SPAN_MIN_WIDTH] [-sp SPAN_PX_PER_STEP]
[-eo EDGE_MAX_OVERLAP] [-el EDGE_MAX_LENGTH]
[-ec EDGE_ANGLE_COST] [-ea EDGE_MAX_ANGLE]
[-f FOCAL_LENGTH] [-z OUTPUT_ZOOM] [-dpi OUTPUT_DPI]
[-nb NO_BINARY] [-s REMAP_DECIMATE]
IMAGE_FILE_OR_FILES [IMAGE_FILE_OR_FILES ...]
positional arguments:
IMAGE_FILE_OR_FILES One or more images to process
optional arguments:
-h, --help show this help message and exit
-d {0,1,2,3}, --debug-level {0,1,2,3}
-o {file,screen,both}, --debug-output {file,screen,both}
-p, --pdf Merge dewarped images into a PDF
-vw SCREEN_MAX_W, --max-screen-width SCREEN_MAX_W
Viewing screen max width (for resizing to screen)
-vh SCREEN_MAX_H, --max-screen-height SCREEN_MAX_H
Viewing screen max height (for resizing to screen)
-x PAGE_MARGIN_X, --x-margin PAGE_MARGIN_X
Reduced px to ignore near L/R edge
-y PAGE_MARGIN_Y, --y-margin PAGE_MARGIN_Y
Reduced px to ignore near T/B edge
-tw TEXT_MIN_WIDTH, --min-text-width TEXT_MIN_WIDTH
Min reduced px width of detected text contour
-th TEXT_MIN_HEIGHT, --min-text-height TEXT_MIN_HEIGHT
Min reduced px height of detected text contour
-ta TEXT_MIN_ASPECT, --min-text-aspect TEXT_MIN_ASPECT
Filter out text contours below this w/h ratio
-tk TEXT_MAX_THICKNESS, --max-text-thickness TEXT_MAX_THICKNESS
Max reduced px thickness of detected text contour
-wz ADAPTIVE_WINSZ, --adaptive-winsz ADAPTIVE_WINSZ
Window size for adaptive threshold in reduced px
-ri RVEC_IDX, --rotation-vec-param-idx RVEC_IDX
Index of rvec in params vector (slice: pair of values)
-ti TVEC_IDX, --translation-vec-param-idx TVEC_IDX
Index of tvec in params vector (slice: pair of values)
-ci CUBIC_IDX, --cubic-slope-param-idx CUBIC_IDX
Index of cubic slopes in params vector (slice: pair of
values)
-sw SPAN_MIN_WIDTH, --min-span-width SPAN_MIN_WIDTH
Minimum reduced px width for span
-sp SPAN_PX_PER_STEP, --span-spacing SPAN_PX_PER_STEP
Reduced px spacing for sampling along spans
-eo EDGE_MAX_OVERLAP, --max-edge-overlap EDGE_MAX_OVERLAP
Max reduced px horiz. overlap of contours in span
-el EDGE_MAX_LENGTH, --max-edge-length EDGE_MAX_LENGTH
Max reduced px length of edge connecting contours
-ec EDGE_ANGLE_COST, --edge-angle-cost EDGE_ANGLE_COST
Cost of angles in edges (tradeoff vs. length)
-ea EDGE_MAX_ANGLE, --max-edge-angle EDGE_MAX_ANGLE
Maximum change in angle allowed between contours
-f FOCAL_LENGTH, --focal-length FOCAL_LENGTH
Normalized focal length of camera
-z OUTPUT_ZOOM, --output-zoom OUTPUT_ZOOM
How much to zoom output relative to *original* image
-dpi OUTPUT_DPI, --output-dpi OUTPUT_DPI
Just affects stated DPI of PNG, not appearance
-nb NO_BINARY, --no-binary NO_BINARY
Disable output conversion to binary thresholded image
-s REMAP_DECIMATE, --shrink REMAP_DECIMATE
Downscaling factor for remapping image
```
- PDF conversion not yet implemented
To try out an example image, run
```sh
git clone https://github.com/lmmx/page-dewarp
cd page-dewarp
mkdir results && cd results
page-dewarp ../example_input/boston_cooking_a.jpg
```
## Explanation and extension to Gpufit
A book on a flat surface can be said to be 'fixed to zero' at the endpoints of a curve, which
you can model as a cubic (see [`derive_cubic.py`](derive_cubic.py))
The "cubic Hermite spline" is one of the models supported by
[Gpufit](https://github.com/gpufit/Gpufit/), a library for Levenberg Marquardt curve fitting in
CUDA (C++ with Python API).
_[Work in progress]_
- See full writeup on [Matt Zucker's blog](https://mzucker.github.io/2016/08/15/page-dewarping.html)
- See [lecture](https://www.cs.cornell.edu/courses/cs4620/2013fa/lectures/16spline-curves.pdf)
on splines by Steve Marschner for more details and how a spline can be represented in matrix form.
- Brief notes on this project are over on [my website](https://doc.spin.systems/page-dewarp)
## Features
Improvements on the original include:
- [x] Banished Python 2
- [x] Command line interface
- [x] Alterable config options
- [x] Repackage for pip installation
- [x] Refactor with modules and classes
- [ ] Speed up the optimisation
- [ ] Multiprocessing on CPU
- [ ] Optional interface to use Gpufit on GPU (or Deep Declarative Networks?)
Raw data
{
"_id": null,
"home_page": "https://github.com/lmmx/page-dewarp",
"name": "page-dewarp",
"maintainer": "",
"docs_url": null,
"requires_python": ">=3",
"maintainer_email": "",
"keywords": "",
"author": "Louis Maddox",
"author_email": "louismmx@gmail.com",
"download_url": "https://files.pythonhosted.org/packages/80/53/39790372c73bb1e4f89f78fb00287604a77cb0f23973fd40a97d8bac1327/page_dewarp-0.1.5.tar.gz",
"platform": null,
"description": "# page-dewarp\n\nDocument image dewarping library using a cubic sheet model\n\nPython 3 library for page dewarping and thresholding,\n[available on PyPI](https://pypi.org/project/page_dewarp/).\n\n## Requirements\n\nPython 3 and NumPy, SciPy, SymPy, Matplotlib, OpenCV, and TOML Kit are required to run `page-dewarp`.\n\n- See [`CONDA_SETUP.md`](https://github.com/lmmx/page-dewarp/blob/master/CONDA_SETUP.md) for\n an example conda environment \n- If you prefer to install everything from pip, run `pip install page-dewarp`\n\nThis library was renovated from the [original (2016) Python 2 script](https://github.com/mzucker/page_dewarp/)\nby Matt Zucker, which you can use if you are somehow still using Python 2.\n\n## Usage\n\n```\nusage: page-dewarp [-h] [-d {0,1,2,3}] [-o {file,screen,both}] [-p]\n [-vw SCREEN_MAX_W] [-vh SCREEN_MAX_H] [-x PAGE_MARGIN_X]\n [-y PAGE_MARGIN_Y] [-tw TEXT_MIN_WIDTH]\n [-th TEXT_MIN_HEIGHT] [-ta TEXT_MIN_ASPECT]\n [-tk TEXT_MAX_THICKNESS] [-wz ADAPTIVE_WINSZ]\n [-ri RVEC_IDX] [-ti TVEC_IDX] [-ci CUBIC_IDX]\n [-sw SPAN_MIN_WIDTH] [-sp SPAN_PX_PER_STEP]\n [-eo EDGE_MAX_OVERLAP] [-el EDGE_MAX_LENGTH]\n [-ec EDGE_ANGLE_COST] [-ea EDGE_MAX_ANGLE]\n [-f FOCAL_LENGTH] [-z OUTPUT_ZOOM] [-dpi OUTPUT_DPI]\n [-nb NO_BINARY] [-s REMAP_DECIMATE]\n IMAGE_FILE_OR_FILES [IMAGE_FILE_OR_FILES ...]\n\npositional arguments:\n IMAGE_FILE_OR_FILES One or more images to process\n\noptional arguments:\n -h, --help show this help message and exit\n -d {0,1,2,3}, --debug-level {0,1,2,3}\n -o {file,screen,both}, --debug-output {file,screen,both}\n -p, --pdf Merge dewarped images into a PDF\n -vw SCREEN_MAX_W, --max-screen-width SCREEN_MAX_W\n Viewing screen max width (for resizing to screen)\n -vh SCREEN_MAX_H, --max-screen-height SCREEN_MAX_H\n Viewing screen max height (for resizing to screen)\n -x PAGE_MARGIN_X, --x-margin PAGE_MARGIN_X\n Reduced px to ignore near L/R edge\n -y PAGE_MARGIN_Y, --y-margin PAGE_MARGIN_Y\n Reduced px to ignore near T/B edge\n -tw TEXT_MIN_WIDTH, --min-text-width TEXT_MIN_WIDTH\n Min reduced px width of detected text contour\n -th TEXT_MIN_HEIGHT, --min-text-height TEXT_MIN_HEIGHT\n Min reduced px height of detected text contour\n -ta TEXT_MIN_ASPECT, --min-text-aspect TEXT_MIN_ASPECT\n Filter out text contours below this w/h ratio\n -tk TEXT_MAX_THICKNESS, --max-text-thickness TEXT_MAX_THICKNESS\n Max reduced px thickness of detected text contour\n -wz ADAPTIVE_WINSZ, --adaptive-winsz ADAPTIVE_WINSZ\n Window size for adaptive threshold in reduced px\n -ri RVEC_IDX, --rotation-vec-param-idx RVEC_IDX\n Index of rvec in params vector (slice: pair of values)\n -ti TVEC_IDX, --translation-vec-param-idx TVEC_IDX\n Index of tvec in params vector (slice: pair of values)\n -ci CUBIC_IDX, --cubic-slope-param-idx CUBIC_IDX\n Index of cubic slopes in params vector (slice: pair of\n values)\n -sw SPAN_MIN_WIDTH, --min-span-width SPAN_MIN_WIDTH\n Minimum reduced px width for span\n -sp SPAN_PX_PER_STEP, --span-spacing SPAN_PX_PER_STEP\n Reduced px spacing for sampling along spans\n -eo EDGE_MAX_OVERLAP, --max-edge-overlap EDGE_MAX_OVERLAP\n Max reduced px horiz. overlap of contours in span\n -el EDGE_MAX_LENGTH, --max-edge-length EDGE_MAX_LENGTH\n Max reduced px length of edge connecting contours\n -ec EDGE_ANGLE_COST, --edge-angle-cost EDGE_ANGLE_COST\n Cost of angles in edges (tradeoff vs. length)\n -ea EDGE_MAX_ANGLE, --max-edge-angle EDGE_MAX_ANGLE\n Maximum change in angle allowed between contours\n -f FOCAL_LENGTH, --focal-length FOCAL_LENGTH\n Normalized focal length of camera\n -z OUTPUT_ZOOM, --output-zoom OUTPUT_ZOOM\n How much to zoom output relative to *original* image\n -dpi OUTPUT_DPI, --output-dpi OUTPUT_DPI\n Just affects stated DPI of PNG, not appearance\n -nb NO_BINARY, --no-binary NO_BINARY\n Disable output conversion to binary thresholded image\n -s REMAP_DECIMATE, --shrink REMAP_DECIMATE\n Downscaling factor for remapping image\n```\n\n- PDF conversion not yet implemented\n\nTo try out an example image, run\n\n```sh\ngit clone https://github.com/lmmx/page-dewarp\ncd page-dewarp\nmkdir results && cd results\npage-dewarp ../example_input/boston_cooking_a.jpg\n```\n\n## Explanation and extension to Gpufit\n\nA book on a flat surface can be said to be 'fixed to zero' at the endpoints of a curve, which\nyou can model as a cubic (see [`derive_cubic.py`](derive_cubic.py))\n\nThe \"cubic Hermite spline\" is one of the models supported by\n[Gpufit](https://github.com/gpufit/Gpufit/), a library for Levenberg Marquardt curve fitting in\nCUDA (C++ with Python API).\n\n_[Work in progress]_\n\n- See full writeup on [Matt Zucker's blog](https://mzucker.github.io/2016/08/15/page-dewarping.html)\n- See [lecture](https://www.cs.cornell.edu/courses/cs4620/2013fa/lectures/16spline-curves.pdf)\n on splines by Steve Marschner for more details and how a spline can be represented in matrix form.\n- Brief notes on this project are over on [my website](https://doc.spin.systems/page-dewarp)\n\n## Features\n\nImprovements on the original include:\n\n- [x] Banished Python 2\n- [x] Command line interface\n - [x] Alterable config options\n- [x] Repackage for pip installation\n- [x] Refactor with modules and classes\n- [ ] Speed up the optimisation\n - [ ] Multiprocessing on CPU\n - [ ] Optional interface to use Gpufit on GPU (or Deep Declarative Networks?)\n\n",
"bugtrack_url": null,
"license": "MIT License",
"summary": "Page dewarping and thresholding using a cubic sheet model.",
"version": "0.1.5",
"split_keywords": [],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "e61871b593c2cd840bd81ce21454025f3db76b683851dabdec283e03f1aa2c41",
"md5": "30d9396e60f785cfeacd74c2c2f080da",
"sha256": "abe293d47ddb5921028773f39957b9e221136ce98fe33d14042cf4354d70adef"
},
"downloads": -1,
"filename": "page_dewarp-0.1.5-py3-none-any.whl",
"has_sig": false,
"md5_digest": "30d9396e60f785cfeacd74c2c2f080da",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3",
"size": 23368,
"upload_time": "2023-04-18T22:14:03",
"upload_time_iso_8601": "2023-04-18T22:14:03.182379Z",
"url": "https://files.pythonhosted.org/packages/e6/18/71b593c2cd840bd81ce21454025f3db76b683851dabdec283e03f1aa2c41/page_dewarp-0.1.5-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "805339790372c73bb1e4f89f78fb00287604a77cb0f23973fd40a97d8bac1327",
"md5": "872ac40efc45e6b3c9fad9d8fffc3cb9",
"sha256": "0d3535c2d7e19fb4732913c400b89661219f96ab4737aebb2a6a91e87c78bb83"
},
"downloads": -1,
"filename": "page_dewarp-0.1.5.tar.gz",
"has_sig": false,
"md5_digest": "872ac40efc45e6b3c9fad9d8fffc3cb9",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3",
"size": 10895533,
"upload_time": "2023-04-18T22:14:08",
"upload_time_iso_8601": "2023-04-18T22:14:08.841012Z",
"url": "https://files.pythonhosted.org/packages/80/53/39790372c73bb1e4f89f78fb00287604a77cb0f23973fd40a97d8bac1327/page_dewarp-0.1.5.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2023-04-18 22:14:08",
"github": true,
"gitlab": false,
"bitbucket": false,
"github_user": "lmmx",
"github_project": "page-dewarp",
"travis_ci": false,
"coveralls": false,
"github_actions": false,
"requirements": [],
"lcname": "page-dewarp"
}
JSON
