YoloV4 weapons Detection

automatic weapon detection software using a convolutional neural network (CNN) based on pre-trained You Only Look Once (YOLO) model.

The software detect three types of weapons : knives, pistols and rifles

deployed locally using Flask

Explore the docs »

View Demo

Table of Contents

1. About The Project
   • Built With
2. Getting Started
   • Prerequisites
   • Installation
3. Acknowledgments

About The Project

The first and supreme goal of science is to serve man-kind, and since the security and safety of man-kind is one of the first and most important priorities, we worked on “Weapon Dectection Sysyem” depending on artificial intelligence and deep learning technologies can detect weapons such as: pistols, knifes and rifles then sending alert mails with the detected weapon for the authorized department such as: police department, security man, ... etc.

(back to top)

Built With

• YoloV4
• Darknet
• Flask
• FFmpeg

(back to top)

Getting Started

To get a local copy up and running follow these simple steps.

Prerequisites

To be able to run the system you have to install the following Prerequisites

• Windows

• Github (Desktop/Bash)

• OpenCV

• Visual Studio community 2019

• CMake GUI: Windows win64-x64

• Cuda

• cuDNN

• Darknet

  git clone https://github.com/AlexeyAB/darknet

• FFmpeg

• Flask

  pip install Flask

Installation

1. Clone the repo

   git clone https://github.com/MahmoudMTaha/YoloV4-Weapons-Detection.git

2. Install Darknet Tutorial in a new folder "darknet-master"
3. delete "data" and "cfg" folders from darknet-master
4. copy "data" and "cfg" folders from the cloned darknet folder to "darknet-master"
5. rename darknet.py in "darknet-master" to darknet2.py
6. download "yolov4-custom_best.weights" from here into "darknet-master"
7. delete the cloned darknet folder
8. rename "darknet-master" to darknet and move it to the project folder
9. edit Yolo_Test_and_Deployment.ipynb by putting sender email and password

(back to top)

Acknowledgments

This is The Graduation Project of the Information Technology institute ITI AI program Powered by EPITA

Team

• Ahmed Magdy - @Linkedin
• Reham ElTagoury - @Linkedin
• Nayrouz Al-Gamaal - @Linkedin
• Mahmoud Taha - @Linkedin - [email protected]
• Wssam Hassan - @Linkedin

(back to top)