CCTV module for Arms and Intention detection and notification

A great deal of Pakistanis have become victims in situations where firearms are involved, such as mass shootings and assaults with weapons. Numerous suggestions have been made as to how to prevent these events from happening. One method that has been used to prevent events like those described above

Project Title

CCTV module for Arms and Intention detection and notification

Project Area of Specialization

Electrical/Electronic Engineering

Project Summary

A great deal of Pakistanis have become victims in situations where firearms are involved, such as mass shootings and assaults with weapons. Numerous suggestions have been made as to how to prevent these events from happening. One method that has been used to prevent events like those described above from happening is surveillance monitoring. Circuit Television Cameras (CCTV’s) are widely used to control occurrence of crimes in the surroundings. Although CCTV’s are deployed at various public and private areas to monitor the surroundings there is no improvement in the control of crimes. This is because CCTV requires human supervision which may lead to human prone errors like missing of some important crime events by human while monitoring so many screens recorded by CCTV’s at same time. To overcome this issue, we came up with Crime Intension Detection System that detects crime in real time videos, images and alerts the human supervisor to take the necessary actions. To alert the supervisors or nearby police station about the occurrence of crime. We added notification system on detection of crime and sending live streaming to concern person whenever crimes are detected. The proposed system is implemented using tensorflow object detection API which detects gun and knife in the hand of person pointing to some other person. 

Project Objectives

The main goal of our project is to prevent mass shootings, active shootings, and mitigate the damage that is a result of these events with an automatic surveillance system. The system has been designed to detect threats (e.g. weapons and knives) in an area, produce an alert when this detection occurs, and generate an organized response to deal with the threat. Our System adds a layer of security when it comes to the prevention of dangerous events, such as active shootings and mass shootings. The system has been designed for object detection capabilities, and behavior detection capabilities in order to create a security system that actively identifies a threat and notifies authorities in real time. The use of our System can help reduce the threat of an attack, such as a mass shooting, as well as assist authorities in capturing criminals. Our System uses Deep Learning (DL) algorithms to analyze multiple data points in order to detect criminal activity faster and more accurately that systems that rely exclusively on human recognition. The system can also be configured to call local law enforcement officials and send them a live video feed of the event. Some of the main features of the system include: being able to detect different types of firearms, recognizing threats in under few seconds, and identifying threats with accuracy. The main goal for our System is to produce accurate and reliable detections of firearms in the drawn state. This is because a drawn firearm will usually indicate a potentially dangerous situation. Firearms that are holstered are usually worn by active law enforcement officials, military personnel, or security personnel. Our Gun Detection System is a real-time threat detection system. The system detects threats, such as firearms, knives, and emotions, through a video feed and sends an alert after the detection.

Project Implementation Method

We are using a Raspberry Pi for our project model. For emotions detection we are using Haar cascade frontal face model to detect faces in live video feed and designed our own little cnn for emotions detection. For firearms and knife detection we are using tensorflow object detection API pre-trained model ssdlite_mobilenet_v2 because it has detection speed of 29ms and mAP of 22.

Benefits of the Project

In Pakistan we have traditional surveillance system installed. Which requires human supervision and human are very prone to errors. Many times a crime active is missed by the person either he is tired of looking at the screens or sleeping. Our system is automated and it won't need any human supervision and will automatically generate an alert when detects firearm and knife. It will prevent mass shooting, bank dacoities etc.

Technical Details of Final Deliverable

We are using a Raspberry Pi for our project model. For emotions detection we are using Haar cascade frontal face model to detect faces in live video feed and designed our own little cnn for emotions detection. For firearms and knife detection we are using tensorflow object detection API pre-trained model ssdlite_mobilenet_v2 because it has detection speed of 29ms and mAP of 22.

Final Deliverable of the Project

HW/SW integrated system

Core Industry

Telecommunication

Other Industries

Security

Core Technology

Others

Other Technologies

Artificial Intelligence(AI)

Sustainable Development Goals

Industry, Innovation and Infrastructure, Sustainable Cities and Communities

Required Resources

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