Loading repository data…

Face-Mask-Detection-Using-YOLOv4 GitHub Details, Stars and Alternatives | OpenRepoFinder

OMEGAMAX10 / repository

Face-Mask-Detection-Using-YOLOv4

Because of the COVID-19 pandemic of 2020, more and more people are concerned with protecting themselves using masks, thus the need of software capable of monitoring whether the people are wearing masks or not. That is why I created a Python application using OpenCV (with CUDA support) based on the YOLOv4 algorithm, capable of monitoring the safety level of a space with video surveillance.

#computer-vision#covid-19#cuda#cuda-support#face-mask-detection#gui

View Repository on GitHub ↗

REPOSITORY OVERVIEW

Live repository statistics

★ 16Stars

⑂ 5Forks

◯ 0Open issues

◉ 16Watchers

47/100

OPENREPOHUB HEALTH SIGNAL

Mixed signals

A transparent discovery signal based on current public GitHub metadata.

Recent activity35% weight

Community adoption25% weight

Maintenance state20% weight

100

License clarity10% weight

100

Project information10% weight

This score does not audit code, security, maintainers, documentation quality, or suitability. Verify the repository and its current documentation before adoption.

README preview

Face Mask Detector using YOLOv4

Because of the COVID-19 pandemic of 2020, more and more people are concerned with protecting themselves using masks, thus the need of software capable of monitoring whether the people are wearing masks or not. That is why I created a PyQt5 application using OpenCV (with CUDA support) based on the YOLOv4 algorithm, capable of monitoring the safety level of a space with video surveillance through multiple video cameras, connected either via USB or network to the system.

Installing dependencies:

For Windows:

Check if the GPU of the system supports CUDA by checking if it is in this list: https://developer.nvidia.com/cuda-gpus
If the GPU supports CUDA, install it using this guide: https://towardsdatascience.com/installing-tensorflow-with-cuda-cudnn-and-gpu-support-on-windows-10-60693e46e781
Install the latest Python 3 version from https://www.python.org/downloads/

Run this command in command line:

PS C:> pip install numpy PyQt5 PyQt5-stubs pyqt5-tools

Install OpenCV with CUDA support using this tutorial: https://medium.com/analytics-vidhya/build-opencv-from-source-with-cuda-for-gpu-access-on-windows-5cd0ce2b9b37

For Ubuntu/Debian:

Check if the GPU of the system supports CUDA by checking if it is in this list: https://developer.nvidia.com/cuda-gpus
If the GPU supports CUDA, install it using this guide: https://medium.com/@pydoni/how-to-install-cuda-11-4-cudnn-8-2-opencv-4-5-on-ubuntu-20-04-65c4aa415a7b
Run these commands:

foo@bar:~$ sudo chmod +x install_dependencies_ubuntu.sh    # make the script executable
foo@bar:~$ sudo ./install_dependencies_ubuntu.sh           # run the script to install the dependencies of the application

Guide for using the Face Mask Detector application:

Execute the Python Script from command line like this:

ALGORITHMICALLY RELATED

Similar Open-Source Projects

Selected from shared topics, language and repository description—not editorial ratings.

OMEGAMAX10 /

Face-Mask-Detector-for-Jetson-Nano

★ 6

Because of the COVID-19 pandemic of 2020, more and more people are concerned with protecting themselves using masks, thus the need of software capable of monitoring whether the people are wearing masks or not. That is why I created a Python application using OpenCV (with CUDA support) based on the YOLOv4 algorithm, more precisely the tiny option, capable of monitoring the safety level of a space with video surveillance. This is the Jetson Nano version of the application.

45/100 health

Active repository

PythonMIT#computer-vision#covid-19#cuda#cuda-support

⑂ 4 forks◯ 0 issuesUpdated Sep 8, 2023

View details View on GitHub ↗

foo@bar:~$ python .\face_mask_detection.py

foo@bar:~$ python3 .\face_mask_detection.py

From the start menu, you can add or delete a camera from the camera list. When creating a camera, a name and an ID must be provided. The ID must be from one of these categories:
- integer (e.g.: 0, 1, 2...): A camera with this ID represents a video recording device physically connected to the system which uses the application. For instance, if you want to use the webcam of a laptop, you must create a camera with an ID of 0 (an explanation would be that, in particular for Ubuntu, the integrated camera of a laptop is interpreted as /dev/video0).
- IP address (e.g.: https://192.168.43.1:8080/video): A camera with this ID represents a video recording device connected to the same network as the system which uses the application. For example, one can connect an Android device as a remote camera using "IP Webcam" Google Playstore app: https://play.google.com/store/apps/details?id=com.pas.webcam&hl=ro&gl=US.
- Video file location: A camera with this ID represents a locally stored video, on which our application will run the detection. This case is useful whenever we have an already pre-recorded video file, possibly from a camera that was not connected to a system with this application.
When you think that the camera list is ready, you can access the main menu.
In the main menu, you will have the following elements:
- A main panel where the selected camera will be displayed together with the detections of the masked or unmasked faces from it, as well as real time statistics of their numbers and its status ("Not Connected" - the camera is not connected to the system, "Safe" - all people wear mask, "Warning" - 1 or 2 people do not wear a mask, "Danger" - more than 3 people do not wear a mask);
- A small selection menu in the upper part of the window from where you can select which camera do you wish to visualize;
- A camera control panel in the right part of the window where the camera list is displayed together with the status of every camera in real time;
- A "Take Photo" button placed in the lower part of the window used for taking photos for further analysis in order to identify the persons not wearing a mask.
Despite the fact that you can take photos manually from the main menu, the application makes each of the connected cameras capable of taking photos automatically whenever its state switches to "Warning" or "Danger".