Skip to content

Built an OCR for parsing Adhar document using pytresseract and opencv image preprocessing

Notifications You must be signed in to change notification settings

Sahana-M/OCR-for-Adhar

Repository files navigation

Actyv ocr : Version 1.0

Usage : This API is designed to extract general fields from Adhar card URL :- https://custom-ocr.herokuapp.com/get_adhar

It accepts 2 types of documents,

  • Adhar front image
  • Adhar back image

Exractions

Adhar front

The general fields extracted from uploaded Adhar front image includes,

Fields extracted
Name
DOB
Gender
Father/Husband name
Age
Adhar number
Document identified

Adhar back

The general fields extracted from uploaded Adhar back image includes,

Fields extracted
Address
Pincode
Document identified

Usage :

Request format
Request - POST
Request URL - https://custom-ocr.herokuapp.com/get_adhar
In Body -> raw format -> <b>JSON request
Example :-
    {
    "file" : "image_url"
    }

Response format - Adhar front
Response type - JSON response
Example :-
{
"age": age,
"dob": "dd/mm/yyyy",
"father": "father name",
"fileIdentified": "document_front.jpg",
"gender": "MALE/FEMALE",
"husband": "husband name",
"idNumber": "adhar number",
"name": "name"
}

Response format - Adhar front
Response type - JSON response
Example :-
{
"address": "xxxxxx",
"fileIdentified" : "document_back.jpg",
"pincode" : "xxxxxx"
}

About

Built an OCR for parsing Adhar document using pytresseract and opencv image preprocessing

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published