Deep Learning Based Automated Visual Surveillance for Hand Gun Detection and Security System

In this project, we are developing a deep learning-based visual surveillance and security system in which a camera will be connected to a system and when a weapon will be displayed in front of it our deep learning model will detect the weapon and will activate the security system including soun

Project Title

Deep Learning Based Automated Visual Surveillance for Hand Gun Detection and Security System

Project Area of Specialization

Electrical/Electronic Engineering

Project Summary

In this project, we are developing a deep learning-based visual surveillance and security system in which a camera will be connected to a system and when a weapon will be displayed in front of it our deep learning model will detect the weapon and will activate the security system including sounding alarms, locking doors and sending a message to police and authorities.

Project Objectives

Apply deep learning algorithms, which rather than substituting the human operator, alert them if a potentially dangerous situation is at hand. This project will be able to detect whether a firearm is present in a frame and if so, it will notify authorities/managers of its detection.

Creating an algorithm that can be integrated into traditional surveillance systems can be used to detect threats faster and more efficiently than those monitored by people.

In modern surveillance systems, there is a person or group of people, in charge of watching monitors which can span across multiple floors of a given area.

Be able to perform real-time detection.

To improve the performance of the object classifier.

Indigenous and cost-effective

Activate bank security systems and close all the sliding doors automatically.

Project Implementation Method

Our project will be implemented on a system in the security room which will be connected to the doors and alarms wirelessly and upon detecting a weapon our system will activate all the security systems.

Benefits of the Project

The users will have an advanced security system that will be efficient in a way that they will not require a person to monitor all security cameras saving the cost of security persons, hence cost-effective, it also eradicates human error which is always possible. This system will also alert the police in the least possible time thus giving less time to robbers to do serious damage.

Technical Details of Final Deliverable

Currently, we are using YOLO v3 algorithm and Deep Neural Network for weapon detection but due to fewer frames on our device, we are also testing some other models which are giving us some promising results. Now, this code will be implemented on a computing device in a security room which will be connected to a controller i.e an Arduino wirelessly. This Arduino will be controlling our doors, alarms, and message sending system which could be an ESP8266 module or a GSM module.

After detecting a weapon our model will send a signal to the Arduino this Arduino will be connected to a motor via H-Bridge which will control the gate, an alarm, esp8266/GSM module, and an electromagnetic lock which will lock the lockers will valuable stuff. The Arduino will activate ale these things and upon resetting the alarms will go off, the electromagnetic lock will disengage and the motor connected to a gate will rotate in reverse thus opening the gate.

Final Deliverable of the Project

HW/SW integrated system

Core Industry

Security

Other Industries

Core Technology

Artificial Intelligence(AI)

Other Technologies

Internet of Things (IoT)

Sustainable Development Goals

Industry, Innovation and Infrastructure

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com