SecureCam - Human Detection and Recording System

Overview

SecureCam is a Python-based application designed to monitor a live camera feed for human presence using OpenCV's face detection. It provides advanced features such as recording videos when a face is detected, real-time FPS display, and user interaction through a GUI built with PyQt5. The app is suitable for use in personal or small-scale security monitoring systems.

Features

• Real-time Face Detection: Detects human faces within a predefined box.
• Auto Recording: Starts recording when a face is detected within the box and stops after a specified duration or if no face is detected for a given time.
• Video Overlay: Displays FPS and timestamp on the video feed.
• GUI Controls:
  • View recordings.
  • Stop recording.
  • View logs.
  • Stop alert.
• Logs System: Tracks and displays system activities.
• Text-to-Speech Integration: Announces recording start and stop events.

Requirements

The following libraries and tools are required to run SecureCam:

• Python 3.8 or later
• OpenCV (cv2)
• PyQt5
• pyttsx3

2. Use the GUI to monitor the camera feed and control the recording system.

GUI Components

• Camera Feed: Displays the live video feed with detection overlays.
• Info Panel: Displays current date, time, and last detection information.
• Logs Panel: Displays a log of system events.
• Control Buttons:
  • View Recordings: Placeholder for viewing recorded videos.
  • Delete Recordings: Placeholder for deleting recordings.
  • View Logs: Displays the system log.
  • Stop Recording: Stops the current recording manually.
  • Stop Alert: Placeholder for stopping alerts.

Recording Behavior

• Recording starts when a face is detected inside the predefined green box.
• Stops if no face is detected for 20 seconds or when the maximum recording duration (10 minutes) is reached.

Key Functions

is_face_in_box(x, y, w, h, box_x, box_y, box_size)

Checks if a detected face lies within the predefined green box.

start_recording(frame, date_time)

Starts recording a video with a unique filename based on the current timestamp.

stop_recording()

Stops the video recording and releases the video writer object.

display_overlay(frame, fps)

Adds an overlay to the video feed showing the current timestamp and FPS.

Customization

• Detection Box: The detection box can be adjusted by modifying box_x, box_y, and box_size.
• Recording Duration: Modify max_recording_duration to change the maximum recording time.
• Alert Messages: Customize the text-to-speech messages in start_recording() and stop_recording().

License

This project is licensed under the MIT License. See the LICENSE file for details.

Acknowledgements

• OpenCV for computer vision tools.
• PyQt5 for GUI development.
• pyttsx3 for text-to-speech functionality.