Intruder Detection - Computer Vision Project

This is a computer vision project developed in Python to detect intruders in a video using a YOLO (You Only Look Once) object detection model. When an intruder is detected within a predefined area, a sound alarm is activated.

Dependencies

To run this project, make sure you have the following libraries installed:

• OpenCV (cv2)
• Ultralytics (ultralytics)
• Pygame (pygame)

You can install the dependencies using the following pip command:

pip install opencv-python ultralytics pygame

Instructions

1. Download the video you want to analyze and save it in the same folder as this script.
2. Run the Python script main.py.
3. The program will start analyzing the video, detecting intruders within the defined area.
4. When an intruder is detected, a sound alarm will be activated.

To terminate the program, press the q key while the video window is in focus.

Notes

• Make sure you have the necessary files for the sound alarm in the same folder as this script.
• The intruder detection area can be adjusted by modifying the values in the area list.
• The input video file is specified in the line video = cv2.VideoCapture('ex01.mp4'). Replace 'ex01.mp4' with the path to your video file.

This project was developed as part of a study in computer vision. If you have any questions or need further assistance, feel free to ask.