Smart Attendance system

• 

  Login Page

• 

  Welcome Page

• 

  Section to upload Students details

• 

  View section to view the uploaded details

• 

  Live attendance Capturing section (Photo is blurred due to privacy concern)

• 

  Section to view Attendance stored and can export in desired format

• 

  Select Dates and View Attendance (Details are hidden due to privacy concern)

• 

  Help section which contains the description of how to use sections

Smart Attendance System

Inspiration

Traditional attendance methods are time-consuming, error-prone, and often allow proxy entries. Inspired by the need for contactless and secure systems, we developed the Smart Attendance System that uses face recognition with fallback 2FA (voice or OTP) to mark attendance accurately and efficiently.

What It Does

• Uses webcam to detect and recognize a person’s face
  
• Marks attendance with name and timestamp
  
• 2FA fallback options if face recognition fails:
  
  • Voice authentication
    
  • OTP via email (SMTP)
    
• Prevents duplicate entries in a session
  
• Stores attendance in a CSV or database
  ##How We Built It
  
• Python – Core logic and ML processing
  
• Flask – Complete web application backend
  
• OpenCV + face_recognition – Face detection and encoding
  
• **SpeechRecognition – Voice-based fallback
  
• SMTP – To send OTPs securely via email
  
• *HTML+CSS *– Web frontend design
  
• CSV & SQLite – Attendance data logging ##Challenges We Ran Into
  
• Handling facial recognition in different lighting and angles
  
• Preventing duplicate attendance entries
  
• Smooth integration between fallback mechanisms
  
• Ensuring OTP delivery using SMTP under time constraints ##Accomplishments That We're Proud Of
  
• Integrated 3 authentication methods in one web app
  
• Successfully developed a fully functional Flask-based attendance system
  
• Built a secure OTP system using SMTP
  
• Achieved high accuracy in face and voice matching
  
• Developed a scalable, user-friendly interface ##What We Learned
  
• Flask web development with ML models
  
• Sending secure OTPs using SMTP and email protocols
  
• Real-time face encoding and recognition
  
• Voice feature extraction and comparison
  
• Optimizing authentication systems for speed and accuracy
  We also applied Euclidean distance for comparing face encodings:

$$ \text{Distance} = \sqrt{\sum_{i=1}^{n} (x_i - y_i)^2} $$

Faces with distance \( < 0.6 \) are considered a match.

What's Next for Smart Attendance System

• Deploy to cloud (AWS/Azure)
  
• Add temperature + mask detection
  
• Mobile app with instant notification
  
• ERP/HRMS integration for institutions and companies
  ##Tagline
  "Look once, get marked — effortless attendance, elevated by AI."

Built With

• cnn
• convolutional
• css
• face-recognition
• flask
• html
• javascript
• network)
• neural
• opencv
• python
• smtp)
• smtplib
• speechrecognition
• sqlite