Introduction

TightOCR provides a thin library to provide an efficient, pleasant, Pythonic interface to Tesseract. Tesseract (https://code.google.com/p/tesseract-ocr/) is the world's most universal OCR project, owned by Google.

The primary goals of this implementation is to provide the following functionalities to Python:

OCR a document and return a block of text. OCR a document, and identify the various parts of a document to allow an application to take advantage of Tesseract's layout analysis.

Secondary functions that are available:

Confidence of recognition HTML-formatted output Slope and margin of text

Though I have tried to provide access to as many of the API methods as possible, there is a very limited amount of documentation available, so many of the more exotic functions haven't been properly tested. IF YOU WANT TO HELP WITH THIS, JUST DO IT AND REGISTER ISSUES OR SUBMIT PULL-REQUESTS.

This library was built as an alternative to python-tesseract (http://code.google.com/p/python-tesseract) for the following reasons:

The usage of SWIG produces an implementation that is excessive and burdensome. python-tesseract is, ironically, incomplete. You are unable to enumerate the parts of the document (getIterator() is broken: http://code.google.com/p/python-tesseract/issues/detail?id=50&can=4&sort=-id)

Requirements

CTesseract (https://github.com/dsoprea/CTesseract) Leptonica (http://code.google.com/p/leptonica/)

Installation

The Leptonica and CTesseract shared-libraries (liblept.so, libctesseract.so) must be findable.

Usage

Return the whole document as text:

from tightocr.adapters.api_adapter import TessApi
from tightocr.adapters.lept_adapter import pix_read

t = TessApi(None, 'eng');

p = pix_read('receipt.png')
t.set_image_pix(p)

t.recognize()
if t.mean_text_confidence() < 85:
    raise Exception("Too much error.")

print(t.get_utf8_text())

Enumerate individual blocks of text (referred to as "paragraphs"), driven by the document's layout:

from tightocr.adapters.api_adapter import TessApi
from tightocr.adapters.lept_adapter import pix_read
from tightocr.constants import RIL_PARA

t = TessApi(None, 'eng');
p = pix_read('receipt.png')
t.set_image_pix(p)
t.recognize()

if t.mean_text_confidence() < 85:
    raise Exception("Too much error.")

for block in t.iterate(RIL_PARA):
    print block