Face Recognition using embedded system

Innovation has always been the driving force of humanity by incorporating new concepts to solve the innumerable problems faced by such as complex structures. Face recognition is an important part of security and surveillance in today's world. The number of thefts and identity fraud has now become a

2025-06-28 16:32:32 - Adil Khan

Project Title

Face Recognition using embedded system

Project Area of Specialization Artificial IntelligenceProject Summary

Innovation has always been the driving force of humanity by incorporating new concepts to solve the innumerable problems faced by such as complex structures. Face recognition is an important part of security and surveillance in today's world. The number of thefts and identity fraud has now become a serious issue. To avoid these thefts and identity fraud, a system for the recognition of the face must be established. The embedded solution will lead to opportunities for unique economic devices that can be used and integrated into most applications because face detection is a computationally intensive task.
There is therefore a need for an efficient and cost-effective system Our purpose is to explore the feasibility of using traditional facial detection and recognition techniques such as Convolutional Neural Network to implement a Raspberry Pi-based facial recognition framework. The objective is to bring face recognition to the level at which the system can replace the use of passwords and RF I-Cards for access to high-security systems and buildings. With the use of the Raspberry Pi kit, we aim to make the system cost-effective and easy to use, with high performance.

Project Objectives

Our main focus is on Face recognition that how it is implemented and by searching on various equipment we will set up this recognition technique onto our embedded system. the main objective of our project is as follows:

Project Implementation Method

Our Project is divided into two main part which is as follows:

  1. Software Part
  2. Hardware Part
Software Part:-

In the software part, we are going to implement our method selected which is Convolution Neural Network through Python Programming. First, we will load our dataset and then preprocess our data so we can get to one convergence point. Next, we will split our data set and then implement CNN on it by passing our set through different layers with filters. Then we will train our dataset and checks it prediction and after implementing several more step move to the hardware portion of our project.

Hardware Part:-

In the hardware portion, we will burn this programming onto our microcontroller which will be Raspberry Pi. along with it, a camera will be attached which will see the face which comes in front of it. The main challenge will be the memory selection that we have to choose a suitable microcontroller that has good speed and enough memory in it that can store code in its memory location.

Add-on (not necessary):-

Along with providing this system, we can apply this system onto our cars with an additional microcontroller which will receive a signal to open the car door locks or not. if not recognized it will not allow the person to open the door locks and if it recognizes then it will send a password to the owner that the car has been unlocked.

Benefits of the Project

The benefits of the project are as follows:-

            With the help of Face Recognition is will be easy to track down any burglars, thieves, or other trespassers. While if it is used on the Government Level it will help to identify terrorists or any other criminals. As for personal use, Face recognition can be used as a security tool for locking personal devices and for personal surveillance cameras.

            Face recognition grant a quick and efficient verification of a person. In addition, it is hard to fill this technology so this is another benefit of this technology.

           Before this technology steps down into existence the security guards had to perform manual identification of a person that took too much time but now this technique is completely independent in the identification process and only takes seconds but is also incredibly accurate. Nowadays the security guard still manually check so we aim to replace this technology permanently so they can guard by just sitting in a room and observing everything from there.

Technical Details of Final Deliverable

Final Deliverable will include the successful Demo of our project in which we will show the working of that device and inside deep knowledge of that how it is implemented. We will present the challenges faced during this project and how we eradicated them to make it convenient for us to get more attention to our project. We will deliver the benefit and objectives behind this project and how to implement this on the market level. we will present our project in the prototype form which will detect the person and recognize them which are present only in the data set provided by us.

Final Deliverable of the Project Hardware SystemCore Industry SecurityOther IndustriesCore Technology OthersOther TechnologiesSustainable Development Goals Industry, Innovation and InfrastructureRequired Resources
Item Name Type No. of Units Per Unit Cost (in Rs) Total (in Rs)
Total in (Rs) 18580
Raspberry PI 4 GB Equipment11600016000
Bluetooth Module Miscellaneous 1400400
Arduino Miscellaneous 1730730
Car actuator Miscellaneous 1750750
Raspberry Pi Camera Equipment1700700

More Posts