Python program to extract text from Aadhar Card

This Python3 program extracts text from a picture of an Aadhar Card.This program 

takes an image (i.e. a jpg or png file) as an argument to the command and validates 

if the image is an Aadhar Card  or not by providing the Aadhar number from the image.

It extracts information from the image like name of the card holder, date of birth

or year of birth ,whichever is available and the Aadhar card number.

import pytesseract
from PIL import Image
import datetime
import cv2
import sys
import os
import os.path
import re
import numpy as np
#class to extract text from an image where the image file is passed as an argument
class Text_Extractor():
    #Constructor
    def __init__(self,image_file):
        self.image_file=image_file
        if self is None:
            return 0
        
#Function to extract the text from image as string 
    def extract_text(self): 

        #img=Image.open(self.image_file)
        img = cv2.imread(self.image_file)
        #resize the image
        img = cv2.resize(img, None, fx=2, fy=2, interpolation=cv2.INTER_CUBIC)
        #convert the image to gray
        img = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
        #the following command uses the tesseract directory path to get the trained data in the config option
        text=pytesseract.image_to_string(img,config='--tessdata-dir "/usr/local/Cellar/tesseract/4.0.0_1/share/tessdata"') 
        return text
    
#class to validate if  an image is a adhar card where the text is passed as an argument
class Aadhar_Card_Validator():
    #Constructor
    def __init__(self,text):
        self.text=text
#Function to validate if an image contains text showing its an aadhar card
    def is_aadhar_card(self):
               res=self.text.split()
               dates={}           
               if 'GOVERNMENT OF INDIA'  in self.text:
                   print ("Aadhar card is valid and the details are below:")
                   index=res.index('INDIA')
                   name=''
                   if res[index+3].isalpha():
                      name= res[index+3] + " " + res[index+4] + " " + res[index+5]
                   else :
                      name= res[index+4] + " " + res[index+5] + " " + res[index+6]
               else:
                    name=res[0] + " " + res[1]
               if len(name)>1:
                   print("Name:  " + name)
               else:
                    print("Name not read")
               p = re.compile('d+/d+/d+')
               if (p.findall(self.text)):
                    dates=p.findall(self.text)
                    
               if len(dates)>0 and len(dates[0])>1:
                   print("Date of birth:"+ str(dates[0]))
               aadhar_number=''
               for word in res:
                  if 'yob' in word.lower():
                       yob=re.findall('d+', word)
                       if yob:
                           print ('Year of Birth: ' + yob[0])
                  if len(word) == 4 and word.isdigit():
                      aadhar_number=aadhar_number  + word + ' '
               if len(aadhar_number)>=14:
                   print("Aadhar number is :"+ aadhar_number)
               else:
                    print("Aadhar number not read")
                    print("Try again or try  another file")


def main():
    
       if len(sys.argv) != 2:
            print ("Wrong number of arguments")
            sys.exit(1)
       image_file_name = sys.argv[1]
         # Check for right infilename extension.
       file_ext = os.path.splitext(image_file_name)[1]
       if file_ext.upper() not in ('.JPG', '.PNG' ):
             print( "Input filename extension should be .JPG or .PNG")
             sys.exit(1)
       te= Text_Extractor(image_file_name)
       text=te.extract_text()
       acv=Aadhar_Card_Validator(text)
       acv.is_aadhar_card()
main()


Input :adhar_card.jpg

Output:Aadhar card is valid and the details are below:
Name:  Kiran Kumari s
Date of birth:01/01/2003
Aadhar number is :4382 5165 5729 

Limitations:The limitations to this program are based on the quality of the image 

passed as input to the program.The text is read by the Optical Character Recognition

(OCR)  functionality of OpenCV provided by tesseract. So in cases where the OCR is 

not done well, the text is either read incorrectly or not read at all.The user might 

want to try different images or change the size or resolution of the image.Another

issue might be unwanted parts in the image, like someone holding the card might have 

the fingers of the person showing.In such cased ,the picture can be cropped and then

used for OCR