A Suspected Person Tracking System Using Cutting Edge Approach The Case Of Snatching Attempts in Shopping Mall
With the help of Internet of Thing (IoT) , Machine Learning (ML) and cloud computing we have proposed a system that identifies a person based on its appearance, gender and location through tracklet generation, re-identification engine and Human machine interface. This can enhance the security o
2025-06-28 16:24:59 - Adil Khan
A Suspected Person Tracking System Using Cutting Edge Approach The Case Of Snatching Attempts in Shopping Mall
Project Area of Specialization Internet of ThingsProject SummaryWith the help of Internet of Thing (IoT) , Machine Learning (ML) and cloud computing we have proposed a system that identifies a person based on its appearance, gender and location through tracklet generation, re-identification engine and Human machine interface. This can enhance the security of location and if any incident took place, i.e., any robbery etc. We can simply identify the current location of thief at any camera and track the path he took by his appearance. Resultantly, it will help our crime investigating departments to lower down crime rate and help people find justice as soon as possible
Project ObjectivesOur Re-Identification (RI) system aim to identify whether a person we're looking for has appeared in another camera at a different moment or even in the same camera at a different time instant.
Objectives of our project are as follow:
- Re-Identify the person of interest
- Determine whether a person of interest has appeared in another camera or not
- Follow the path a person of interest took
- Enhance security of place
- Make sure no illegal activities occurring
- Make re-identification process less time consuming and more accurate.
Our proposed system will identify a person based on its appearance and gender by using machine learning algorithms such as object detection API, gender detection and person segmentation. First of all, we will say out about the physical appearance of required person through our voice input module in URDU, then that information will be translated in English through Google API, further the Input added through voice and video recorded will be collected by Raspberry PI and it will send them to online cloud platform called GCP. At GCP different algorithms will be run and finally the information of required person will be displayed. After that we can simply identify the current location of thief at any camera and track the path he took by his/her appearance.
Benefits of the ProjectIf we are talking about Potential applications of this project, it is significant that it is very effective in dealing with modern problems faced by typically used CCTV cameras i.e., it’s hard to identify and keep track on any particular person in a crowd but due to this project we can keep a proper eye on any specific person out there in crowd though his appearance, age, gender etc. In this world with full of crimes it’s really important to keep eye on everything happening around us and hence keeping balance. So, in that case this project is playing a valid role in finding and tracking any specific person in a crowd.
Technical Details of Final DeliverableFor prototype, two cameras will be placed at non over lapping place, Cameras and microphone will be connected to Raspberry PI. Raspberry PI will be connected to stable internet connection. System admin will say out about the physical appearance of required person i.e his upper body dress color and lower body dress color or his gender through our voice input module in URDU, then that information will be translated in English through Google API, further the Input added through voice and video recorded will be collected by Raspberry PI and it will send them to online cloud platform called GCP. At GCP different algorithms will be run i.e object detection and person with similar upper body and lower body color dress or his different attributes will be detected and information of required person will be displayed. After that we can simply identify the current location of thief at any camera and track the path he took by his/her appearance.
It will enhance security of the place.
| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 44500 | |||
| Raspberry Pie | Equipment | 1 | 23500 | 23500 |
| CCTV Cameras | Equipment | 2 | 7000 | 14000 |
| Wires For Connectivity | Equipment | 1 | 4500 | 4500 |
| Microphone | Equipment | 1 | 2500 | 2500 |