Gunshot Source Detection and Localization
Our country has been in the state of war for the last two decades. The armed forces needed advance solutions to efficiently tackle and combat this war on terror. With the advancement in technology, intelligent systems can be built to assist Pakistan army. In this project, our aim is to automate the
2025-06-28 16:32:49 - Adil Khan
Gunshot Source Detection and Localization
Project Area of Specialization Artificial IntelligenceProject SummaryOur country has been in the state of war for the last two decades. The armed forces needed advance solutions to efficiently tackle and combat this war on terror. With the advancement in technology, intelligent systems can be built to assist Pakistan army. In this project, our aim is to automate the process of gunshot detection using deep-learning model. After detection of sound, we will also use array signal processing algorithm to localize the source of sound to provide azimuth and elevation of the gun. For this purpose, a motor assembly will be utilized which rotate the gun in the direction of source.
Project ObjectivesMain Objective:
The main objective of this project is to provide efficient and automated surveillance to our security agencies.
Sub-Objectives:
- Build a system that efficiently detects, identifies and differentiates the gunshot sound with in noise (sounds other than gunfire)
- Build a system which after detection is also capable of its localization.
Our system is consisted of following three parts.
- Training
- Testing
- Localisation
In first part, the system is trained using training techniques and algorithms. Afterwards this trained model distinguishes the sound of gunfire from other sounds (referred as noise). After that it will make a decision that this sound is of gunshot or any other. On the basis of that decision, if it is a gunshot then it will drive the location of that sound by the help of azimuth and elevation and display it on screen.
Benefits of the ProjectThe project provides the following benefits.
- Effective surveillance in metropolitans and in border areas.
- Can be used to allocate the hidden enemies in critical areas.
- Automated systems that can be effectively reduce the crime ratios in cities.
Actually what we have done and planned to submitt on the given time is that first we have made the spectrogram of the collected sounds, then we run the fastAI algorithms in order to train our model according to the collected sounds.Fastai is the first deep learning library to provide a single consistent interface to all the most commonly used deep learning applications for vision, text, tabular data, time series, and collaborative filtering.Then comes the testing part in which we will test our trained model with different sounds and live sound.Then comes the localization part in which we will try to localize the gun according to the sound intercepted.
Final Deliverable of the Project HW/SW integrated systemCore Industry SecurityOther Industries Others Core Technology Artificial Intelligence(AI)Other Technologies Artificial Intelligence(AI)Sustainable Development Goals Decent Work and Economic Growth, Industry, Innovation and Infrastructure, Peace and Justice Strong Institutions, Partnerships to achieve the GoalRequired Resources| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 73286 | |||
| Roland OctaCapture | Equipment | 1 | 69536 | 69536 |
| Proposal+Mid+Final Report Printing | Miscellaneous | 3 | 100 | 300 |
| Standee For Open House | Miscellaneous | 1 | 250 | 250 |
| Poster Of FYP For Standee | Miscellaneous | 1 | 200 | 200 |
| Thesis Book | Miscellaneous | 3 | 1000 | 3000 |