Behavior Detection System for Autistic Kids

Introduction

This project aims to assist autism centers in monitoring and improving student behavior through a machine learning-based behavior detection system. The solution includes behavior classification, student-specific introductory videos, and detailed behavior analysis.

---

Key Features

1. Behavior Detection:

   • Classifies 8 behaviors:
     • Hand flapping
     • Head beating
     • Biting
     • Sitting in chair (active)
     • Sitting in chair (inactive)
     • Running
     • Fighting
   • Tracks behavior duration and generates a pie chart for analysis.

2. Personalized Student Experience:

   • Face recognition for student identification.
   • Plays an introductory video unique to each student after detecting their presence.

3. Classroom Monitoring:

   • Webcam-based real-time face recognition and behavior analysis.
   • Records the time spent on each behavior for detailed tracking.

4. Face Attendance System:

   • Tracks student attendance via face recognition.
   • Plays advertisements or other default content when no student is detected.

---

System Design

Architecture

1. Frontend: Built using Tkinter for a simple and user-friendly UI.
2. Backend:
   • Custom-trained Machine Learning model for behavior classification.
   • Python-based scripts for integration with hardware (webcam).

Hardware Requirements

• Webcam
• Computer with moderate processing capabilities

Software Requirements

• Python (Version 3.7+)
• Required libraries:

  pip install opencv-python mediapipe scikit-learn matplotlib numpy

---

Project Workflow

1. Face Attendance System

• Starts when the system detects a student using a webcam.
• If a registered student is identified:
  • Stops playing the advertisement.
  • Plays the student’s introductory video.

2. Behavior Detection

• The classroom webcam monitors students and identifies their behaviors.
• The behavior classification system tracks each student’s activities in real-time.

3. Behavior Analysis

• Generates a pie chart summarizing each student’s behavior distribution.
• Provides insights into behavioral improvements over time.

---

Code Snippets

Face Detection and Attendance

import cv2
import face_recognition

# Initialize the webcam
video_capture = cv2.VideoCapture(0)

while True:
    ret, frame = video_capture.read()
    rgb_frame = frame[:, :, ::-1]
    face_locations = face_recognition.face_locations(rgb_frame)
    
    for top, right, bottom, left in face_locations:
        cv2.rectangle(frame, (left, top), (right, bottom), (0, 255, 0), 2)
    
    cv2.imshow("Attendance System", frame)
    
    if cv2.waitKey(1) & 0xFF == ord('q'):
        break

video_capture.release()
cv2.destroyAllWindows()




### run gui.py