Aadhaar OCR API

A lightweight Python project that extracts key information from the front and back images of an Indian Aadhaar card using Tesseract OCR and exposes it through a simple Flask REST API.

---

🔍 What it does

- Uses image-to-text OCR with `pytesseract`
- Parses:
  - ✅ Name
  - ✅ Date of Birth
  - ✅ Gender
  - ✅ Aadhaar Number
  - ✅ Parent/Guardian Name & Relation
  - ✅ Address
  - ✅ Pincode
- Accepts images through an API or local file path

---

📁 Project Structure

├── aadhar_ocr.py # Core OCR logic (modularized functions)
├── app.py # Flask API version using uploaded images
├── test.py # Flask-RESTful variant (same logic, different structure)
├── file1.py # Initial version with local image path + file output
├── extracted_aadhar_data.json # Sample output from file1.py
├── extracted_back_data.json # Sample output from file1.py

---

🛠️ Requirements

• Python 3.7+
• Tesseract OCR installed (add to PATH)
• Python libraries:
• bash
• pip install flask flask-restful pytesseract numpy regex

To Run the files, there's different versions.

Initially, install all the libraries using the requirements.txt file. If you don't know how to, just write this command.

pip install -r requirements.txt

1. Using Flask Rest API

a. Here, you go to postman or any suitable provider and then give an appropriate POST REQUEST for Front Image and Back Image # POST http://localhost:5000/ # To run the code for this, just type "python test.py"

b. Sample response is as:

  {
  "Name": "Sid Malhorta",
  "DOB": "28-05-2000",
  "Gender": "Female",
  "Aadhaar": "3425 0653 1151",
  "Parent": "Amit Kumar Tandon",
  "Relation": "S/O",
  "Address": "Merlin Greens, Flat A 1920, Kriparampur, Naskar Para, Chandi, Bishnupur, South 24 Parganas, West Bengal",
  "Pincode": "743503"
  }
  

2. Using Flask API

Do the above and just type "python app.py"

a. Sample response is as:

  {
  "Name": "Sid Malhorta",
  "DOB": "28-05-2000",
  "Gender": "Female",
  "Aadhaar": "3425 0653 1151",
  "Parent": "Amit Kumar Tandon",
  "Relation": "S/O",
  "Address": "Merlin Greens, Flat A 1920, Kriparampur, Naskar Para, Chandi, Bishnupur, South 24 Parganas, West Bengal",
  "Pincode": "743503"
  }
  

3. For running locally

a. Go to the file file1.py b. Give paths for Front Image and Back Image # Now just type "python file1.py" c. The extracted data will be stored in the form of JSON files in your system. b. Sample response is as:

Front file:

     {
      "Name": "Sid Malhorta",
      "DOB": "28-05-2000",
      "Gender": "Female",
      "Aadhaar": "3425 0653 1151"
     }
   

Back file:

     {
    "Parent": "Amit Kumar Tandon",
    "Relation": "S/O",
    "Address": "Merlin Greens, Flat A 1920, Kriparampur, Naskar Para, Chandi, Bishnupur, South 24 Parganas, West Bengal",
    "Pincode": "743503"
    }
   

---

Summary

File	Description
file1.py	Initial Aadhaar parser using local images and writing JSON output
aadhar_ocr.py	Core reusable OCR logic now used by both app.py and test.py
app.py	Simple Flask API that receives image files via POST
test.py	Enhanced Flask-RESTful version with better structure and error handling