search

afrozchakure / Aadhar-OCR

This is a repository for Aadhar OCR 💳
GNU General Public License v3.0
6 stars 2 forks source link

Available models for YOLOv3 and Faster RCNN #3

Open afrozchakure opened 4 years ago

afrozchakure commented 4 years ago

OCR with Tesseract: https://nanonets.com/blog/ocr-with-tesseract/#ocrwithpytesseractandopencv

Steps followed:

  1. Convert to Grayscale.
  2. Threshold.
  3. Find Contours.
  4. Draw BoundingRect.

Faster RCNN - (The link to the Github repo provided by you)

  1. The above is a card detector model only and doesn't extract any information from our Image.
  2. We would need to build our own card information extraction model.

id_detector

What I found:

Resource for Aadhar Card Detection and Info Extraction

Link : https://github.com/dilippuri/Aadhaar-Card-OCR
(Works well for implementation :point_up_2: :point_up_2: purposes)

Note: For the above resource the we need to have a list of people's names for the model to recognize it and store in a JSON format file.