This package contains a tool for automatically cropping and deskewing
images of book pages captured by an Internet Archive Scribe bookscanner.

This software was written as a side project in 2008 and has not been
optimized. In its current state, it can process a single page image in
about 2 seconds on a Scribe bookscanner. Since it is too slow to run in
realtime during the scanning process, it was never deployed.
(Update: it was deployed in 2011)


autoCropFoldout.c contains code that crops to the edges of a page shot
against a black background. If you are investigating this project for
your own scanning, autoCropFoldout would probably be the best place to
start.

Example output of autoCropFoldout is available here:
http://archive.org/download/autocrop_test_data/foldout_test.zip/foldout_test%2Findex.html


autoCropScibe.c contains the autocrop code for the Scribe bookscanner.

Example output of the autocrop algorithm is available here:
http://www.archive.org/download/autocrop_test_picturesquenewen00swee/picturesquenewen00swee/index.html

processScribe.py is a wrapper that will process images after the image
capture phase is complete and will write crop and skew information into
scandata.xml (which is created by Scribe).

autoCropScribe depends on the Leptonica image processing library. This
tool has been compiled against Leptonica 1.56 and has not been tested
with more modern versions of Leptionica.
(Update: now compiled with Leptonica 1.68)


To compile autocrop scribe simply run:

$ make


To run tests, compile with `make CXXFLAGS=-DWRITE_DEBUG_IMAGES`

Here is a walkthrough for building, downloading test data, and running tests:

# install dependencies
$ sudo apt-get install libjpeg-dev libpng-dev libtiff-dev

# clone and build
$ git clone [email protected]:rajbot/autocrop.git
$ cd autocrop/
$ make CXXFLAGS=-DWRITE_DEBUG_IMAGES

# download test images and run tests
$ cd tests/
$ wget http://archive.org/download/autocrop_test_data/autocrop_test_data_images.zip
$ unzip autocrop_test_data_images.zip
$ mv autocrop_test_data_images testimg
$ wget http://archive.org/download/autocrop_test_data/testimg_foldout.zip
$ unzip testimg_foldout.zip
$ make test