This is a python model for extracting info from id
Problem:
Extract information from image of RCI License by OCR in proper format.
Information like -
Name, Year of Birth, Gender, UID
Solution:
Steps:
-> Take image
-> crop to box(which has text in it)
-> convert into gray scale(mono crome)
-> give to tesseract
-> text(output of tesseract)
Now we will process this text means we will get meaningful information from it.
-> find name
-> find phone number
-> find for RCI Id
then we will validate the data with the info provided by the user, and the license.
Dependent packages
use the command '''pip install -r requirements.txt''' then run '''uvicorn app:app --reload'''
Structure and Usage
Directories:
prediction_service -
which contains code files for prediction service
static -
place to store static file
app.py -
fast_api file
Usage:
python file_name.py [input image]
Output will be JSON object name
💯