kraken

OCR engine compatible with ocropus

Project description

Description

https://travis-ci.org/mittagessen/kraken.svg

kraken is a fork of ocropus intended to rectify a number of issues while preserving (mostly) functional equivalence. Its main goals are:

Explicit input/output handling ✓

Clean public API

Word bounding boxes in hOCR ✓

Tests

Removal of runtime dependency on gcc ✓

Removal of unused spaghetti code ✓

Ticked of goals have been realized while some others still require further work. Pull requests and code contributions are always welcome.

Installation

While kraken does not require a working C compiler on run-time anymore numpy and scipy compilation still requires build-essential or your distributions equivalent. Because the build behavior of pip versions older than 6.1.0 interferes with the scipy build process numpy has to be installed before doing the actual install:

# pip install numpy

Install kraken either from pypi:

$ pip install kraken

or by running pip in the git repository:

$ pip install .

Finally you’ll have to scrounge up an RNN to do the actual recognition of characters. To download ocropus’ default RNN and place it in the kraken directory for the current user:

$ kraken download

Quickstart

To binarize a single image using the nlbin algorithm:

$ kraken binarize grey.png bw.png

To segment a binarized image into reading-order sorted lines:

$ kraken segment bw.png lines.txt

To OCR a binarized image using the default RNN and the previously generated page segmentation:

$ kraken ocr --lines lines.txt bw.png

Documentation

Have a look at the documentation

Project details

Release history Release notifications | RSS feed

5.3.0

Nov 21, 2024

5.2.9

Aug 27, 2024

5.2.8

Jul 24, 2024

5.2.7

Jul 8, 2024

5.2.6

Jul 3, 2024

5.2.5

May 23, 2024

5.2.4

May 9, 2024

5.2.3

May 5, 2024

5.2.2

Apr 30, 2024

5.2.1

Apr 22, 2024

5.2.0

Apr 20, 2024

5.0.0 yanked

Mar 28, 2024

Reason this release was yanked:

broken polygonization

4.3.13

Jul 19, 2023

4.3.12

May 12, 2023

4.3.11

Apr 21, 2023

4.3.10

Apr 17, 2023

4.3.9

Mar 20, 2023

4.3.7

Mar 6, 2023

4.3.6

Feb 23, 2023

4.3.5

Feb 21, 2023

4.3.4

Feb 20, 2023

4.3.3

Feb 14, 2023

4.3.2

Feb 14, 2023

4.3.1

Feb 14, 2023

4.3.0

Feb 13, 2023

4.2.0

Aug 29, 2022

4.1.2

Apr 12, 2022

4.1.1

Apr 11, 2022

4.1.0

Apr 5, 2022

4.0.0

Feb 22, 2022

3.0.13

Apr 12, 2022

3.0.9

Feb 22, 2022

3.0.8

Feb 3, 2022

3.0.7

Jan 24, 2022

3.0.6

Nov 7, 2021

3.0.4

Jul 28, 2021

3.0.2

Jun 28, 2021

3.0.1

Jun 25, 2021

3.0.0.0b25 pre-release

May 26, 2021

3.0.0.0b24 pre-release

Apr 22, 2021

3.0.0.0b23 pre-release

Mar 18, 2021

3.0.0.0b22 pre-release

Feb 22, 2021

3.0.0.0b21 pre-release

Feb 17, 2021

3.0.0.0b20 pre-release

Feb 11, 2021

3.0.0.0b20.dev7 pre-release

Feb 3, 2021

3.0.0.0b19 pre-release

Jan 7, 2021

2.0.8

Nov 25, 2019

2.0.5

May 14, 2019

2.0.4

May 14, 2019

2.0.3

May 14, 2019

2.0.2

May 14, 2019

2.0.1

Feb 25, 2019

2.0.0

Feb 25, 2019

1.0.1

Dec 11, 2018

1.0.0

Dec 10, 2018

0.9.16 yanked

Apr 20, 2018

0.9.15 yanked

Apr 18, 2018

0.9.14 yanked

Apr 18, 2018

0.9.13 yanked

Apr 17, 2018

0.9.12 yanked

Apr 17, 2018

0.9.11 yanked

Apr 17, 2018

0.9.10 yanked

Mar 2, 2018

0.9.9 yanked

Mar 2, 2018

0.9.8 yanked

Jan 8, 2018

0.9.7 yanked

Nov 1, 2017

0.9.6 yanked

Oct 31, 2017

Reason this release was yanked:

conflict with modern install tools

0.9.4 yanked

Jul 19, 2017

0.9.3 yanked

Jul 18, 2017

0.9.2 yanked

May 15, 2017

0.9.0 yanked

Nov 3, 2016

0.7.6 yanked

Jan 18, 2016

0.7.5 yanked

Dec 14, 2015

0.7.4 yanked

Nov 25, 2015

0.7.3 yanked

Nov 3, 2015

0.7.2 yanked

Oct 18, 2015

0.7.1 yanked

Sep 22, 2015

0.7.0 yanked

Sep 17, 2015

0.6.3 yanked

Sep 14, 2015

0.6.2 yanked

Sep 12, 2015

0.6.2.dev1 pre-release yanked

Sep 12, 2015

0.5.0 yanked

Sep 7, 2015

0.4.7 yanked

Aug 26, 2015

0.4.6 yanked

Aug 13, 2015

0.4.5 yanked

Jul 16, 2015

0.4.5.dev1 pre-release yanked

Aug 13, 2015

This version

0.4.4 yanked

Jun 16, 2015

0.4.3 yanked

Jun 14, 2015

0.4.2 yanked

May 30, 2015

0.4.1 yanked

May 26, 2015

0.3.4 yanked

May 23, 2015

0.3.3 yanked

May 22, 2015

0.3.1.post11 yanked

May 22, 2015

0.3.1.post10 yanked

May 22, 2015

0.3.1 yanked

Apr 23, 2015

0.2.5 yanked

Apr 16, 2015

0.2.4 yanked

Apr 16, 2015

0.2.3 yanked

Apr 16, 2015

0.2.2 yanked

Apr 5, 2015

0.1.0 yanked

Mar 30, 2015

Reason this release was yanked:

conflict with modern install tools

0.1-dev pre-release yanked

Dec 16, 2013

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

kraken-0.4.4.tar.gz (493.6 kB view details)

Uploaded Jun 16, 2015 Source

Built Distribution

kraken-0.4.4-py2.py3-none-any.whl (25.6 kB view details)

Uploaded Jun 16, 2015 Python 2 Python 3

File details

Details for the file kraken-0.4.4.tar.gz.

File metadata

Download URL: kraken-0.4.4.tar.gz
Upload date: Jun 16, 2015
Size: 493.6 kB
Tags: Source
Uploaded using Trusted Publishing? No

File hashes

Hashes for kraken-0.4.4.tar.gz
Algorithm	Hash digest
SHA256	`bae0706c7f215bf2a696042fde2de171c757a7ce1aed58cabeec2ab381fcf18d`
MD5	`d582822d92e0db94a420e35a2f9427b8`
BLAKE2b-256	`77209a2a12a6baae422fa6b00161b7248c138ac2268ca737c102d5dd96b504bc`

See more details on using hashes here.

File details

Details for the file kraken-0.4.4-py2.py3-none-any.whl.

File metadata

Download URL: kraken-0.4.4-py2.py3-none-any.whl
Upload date: Jun 16, 2015
Size: 25.6 kB
Tags: Python 2, Python 3
Uploaded using Trusted Publishing? No

File hashes

Hashes for kraken-0.4.4-py2.py3-none-any.whl
Algorithm	Hash digest
SHA256	`9ff13ad8846de881479826259dde8baf086cf18d507b8bbfdf7d695e17661a96`
MD5	`07251d28730476407ba8c74b7acb3156`
BLAKE2b-256	`c6f892fbd7174309ddc54e4f2ea203893ac24d2fd8a07f3c0818a4bf134b3b84`