MCQ Scanner — AI-Powered Quiz Digitizer

An AI-powered mobile and backend system that automates the conversion of printed or handwritten multiple-choice question (MCQ) papers into interactive, digital quizzes. The project bridges the gap between traditional paper-based assessments and modern digital learning environments.

---

Features

• MCQ Paper Scanning
  Upload or capture handwritten/printed MCQ papers through the mobile app.

• AI Text Extraction
  Extracts question statements and options using Google Gemini API.

• Automatic Quiz Structuring
  Converts scanned text into structured, digital quiz format.

• Quiz Evaluation
  Enables users to take quizzes and view results instantly.

• Quiz History Tracking
  Stores and retrieves user quiz results.

• Offline Fallback
  Uses a rule-based parser if Gemini AI is unavailable.

---

Screenshots

Splash Screen	Upload Screen	Quiz Generated
Select Answer	Evaluation Report	History

---

Tech Stack

• Frontend: Flutter ^3.9.2
• State Management: GetX ^4.7.2
• Backend: Flask ^3.0.0
• AI Model: Google Gemini (gemini-2.5-flash)
• Database: SQLite (for backend history) and SharedPreferences (for app cache)

---

Project Structure

mcq_scanner/
├── mcq_scanner_app/              # Flutter mobile application
│   ├── lib/
│   │   ├── core/                 # Constants, themes, routes
│   │   ├── data/                 # Models
│   │   ├── modules/              # Screens (Home, Scan, Quiz, etc.)
│   │   └── services/             # OCR and API services
│   └── assets/
│       └── images/app-ui/        # App screenshots
└── mcq_scanner_backend/          # Flask backend server
    ├── app/
    │   ├── routes.py             # API endpoints
    │   ├── services/             # AI processing logic
    │   └── models/               # Database operations
    └── run.py                    # Server entry point

---

Getting Started

Prerequisites

• Flutter SDK ^3.9.2
• Python ^3.10
• Google Gemini API Key

---

Installation & Setup

1. Backend Setup (Flask)

1. Navigate to the backend folder:

   cd mcq_scanner_backend

2. Create a virtual environment:

   python -m venv venv
   source venv/bin/activate    # For Linux/Mac
   venv\Scripts\activate       # For Windows

3. Install dependencies:

   pip install -r requirements.txt

4. Configure environment variables:

   • Copy .env.example to .env
   • Add your GEMINI_API_KEY

5. Run the server:

   python run.py

2. Frontend Setup (Flutter)

1. Navigate to the app folder:

   cd mcq_scanner_app

2. Install dependencies:

   flutter pub get

3. Configure API URL:

   • Update the baseUrl in lib/core/constants/app_constants.dart with your machine's local IP address.

4. Run the app:

   flutter run

---

API Endpoints

Endpoint	Method	Description
/api/process_text	POST	Extracts questions and options from raw text.
/api/evaluate_quiz	POST	Evaluates quiz answers using AI or self-assessment.
/api/quiz_history	GET	Retrieves user's previous quiz attempts.

---

Dependencies

Flutter App

dependencies:
  get: ^4.7.2
  google_mlkit_text_recognition: ^0.15.0
  image_picker: ^1.2.0
  http: ^1.5.0
  camera: ^0.11.2

Flask Backend

Flask==3.0.0
python-dotenv==1.0.0
google-generativeai==0.3.2

---

Author

Abdul Sami

• GitHub: @5-abdulsami
• Website: abdulsami.live
• Email: 5abdulsami2004@gmail.com
• LinkedIn: Abdul Sami