Mask Detection | Call for Code 2020

This project was developed for the IBM Call For Code hackathon, 2020

📋 Table of contents

• Introduction
• The idea
• Demonstration
• Running Locally
• Technologies
• Back-end
• Contributors
• Author
• License

📯 Introduction

The pandemic generated by the new coronavirus has caused numerous impacts on society. As a way of minimizing the impacts, we created a solution that aims to help the owners of establishments to have an idea of ​​how many people are in each region within the place and if some of these people are not wearing masks.

💡 The idea

With the help of the establishment's own security cameras, we used some image recognition tools within the IBM Cloud to analyze whether or not the person is wearing masks and in addition, we created a heat map of the region to show which places have a greater amount of people. This information can be useful if those responsible for the establishment want to reorganize the products in order to distribute people between different regions.

🖥️ Demonstration

If you want, watch the video of the demo in operation: https://www.youtube.com/watch?v=tbppCvWFpVQ&rel=0

Home screen

On the home screen, the footage available for the responsible person to view is displayed.

Detections screen

On this screen, it is possible to view the video with the system by detecting people with and without masks. With this information and hands, we have a button that sends an audible alert on the establishment's sound system, informing the importance of wearing masks.

Below we have a heat map of the establishment, showing the region that the security camera can reach and a color that varies between green and red according to the number of people in that region. This information can be important for the owner of the establishment, since he can reorganize the products in such a way as to make that region not stay with so many people and consequently with the green color.

Camera heat map

Heat map indicating the risk of contagion in the region of the camera.

🕹️ Running locally

Prerequisites

1. Installing Git: You need to have Git on your machine to perform a few steps. To download Git, click here.


2. Installing Yarn 1: Yarn is a package manager that you can download directly from the website by clicking here. If you prefer, use the NPM.


3. Code Editor (optional): Make sure you have a code editor of your choice. I recommend using the VS Code. If you need to, download it here.

Running the App

1. In a terminal, clone this repository:

   git clone https://github.com/Silvio-Ronaldo/DetectorDePessoasComMascaras.git

2. Enter the project folder:

   cd DetectorDePessoasComMascaras

3. Install all dependencies:

   yarn install or yarn

4. Start the development server:

   yarn start

5. After these steps, the server should start at the 3000 port, open the browser and access http://localhost:3000.

🛡️ Technologies

The following tools were used in the development of the project:

• Text To Speech
• IBM Cloud Anottations
• Watson Machine Learning

⚙️ Back-end

This is the Back-end link: https://call4code-detect-mask.herokuapp.com/

🤝 Contributors

Joed Silva ☕

Dorival ☕

👽 Author

Silvio Ronaldo 🍀

Leave your star, fork the project or open a pull request ❤️

Contact me on social networks:

⚖️ License

Mask Detection is MIT licensed, as found in the LICENSE file.