Skip to main content

Meaningful Optical Character Recognition from identity cards with Deep Learning.

Project description


Meaningful Optical Character Recognition from identity cards with Deep Learning.


mocr is a library that can be used to detect meaningful optical characters from identity cards. Code base is pure Python and works with 3.x versions. It has some low level dependencies such as Tesseract. mocr uses a pre-trained east detector with OpenCV and applies it’s Deep Learning techniques.

It has a pre-trained east detector inside the module and a custom trained model can be given as a parameter.


  • Tessaract must be installed on your computer before using OCR. Please check installation link for details.
  • The other dependencies are listed on requirements.txt and will be installed when you install with pip.


From source

Install module using pip:

$ pip install mocr

Download the latest mocr library from:

Install module using pip:

$ pip install -e .

Extract the source distribution and run:

$ python build
$ python install

Running Tests

The test suite can be run against a single Python version which requires pip install pytest and optionally pip install pytest-cov (these are included if you have installed dependencies from requirements.testing.txt)

To run the unit tests with a single Python version:

$ py.test -v

to also run code coverage:

$ py.test -v --cov-report html --cov=mocr

To run the unit tests against a set of Python versions:

$ tox

Sample Usage

  • text_recognition Initiating the TextRecognizer with identity image and then finding the texts with their frames:
import os
from mocr import TextRecognizer

image_path = os.path.join('tests', 'data/sample_uk_identity_card.png')
east_path = os.path.join('mocr', 'model/frozen_east_text_detection.pb')

text_recognizer = TextRecognizer(image_path, east_path)
(image, _, _) = text_recognizer.load_image()
(resized_image, ratio_height, ratio_width, _, _) = text_recognizer.resize_image(image, 320, 320)
(scores, geometry) = text_recognizer.geometry_score(east_path, resized_image)
boxes = text_recognizer.boxes(scores, geometry)
results = text_recognizer.get_results(boxes, image, ratio_height, ratio_width)

# results: Meaningful texts with bounding boxes
  • face_detection:
from mocr import face_detection

face_image = face_detection.detect_face(image_path)
# face_image is the byte array detected and cropped image from original image
from mocr import face_detection

face_image = face_detection.detect_face_from_video(video_path)
# face_image is the byte array detected and cropped image from original video


Sample command line usage

  • Optical Character Recognition
python -m mocr --image tests/data/sample_uk_identity_card.png --east tests/model/frozen_east_text_detection.pb
  • Face detection from image file
python -m mocr --image-face 'tests/data/sample_de_identity_card.jpg'
  • Face detection from video file
python -m mocr --video-face 'tests/data/face-demographics-walking.mp4'






Project details

Download files

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

Files for mocr, version 0.4.0
Filename, size File type Python version Upload date Hashes
Filename, size mocr-0.4.0-py3-none-any.whl (146.4 kB) File type Wheel Python version py3 Upload date Hashes View

Supported by

AWS AWS Cloud computing Datadog Datadog Monitoring DigiCert DigiCert EV certificate Facebook / Instagram Facebook / Instagram PSF Sponsor Fastly Fastly CDN Google Google Object Storage and Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Salesforce Salesforce PSF Sponsor Sentry Sentry Error logging StatusPage StatusPage Status page