Face Detection and Recognition Using Raspberry Pi

Abstract

The face detection and recognition system using Raspberry Pi is a security system that uses advanced algorithms to detect and recognize faces. The system is designed to authorize access to individuals with authorized faces and deny access to unauthorized individuals. The system integrates with a door lock, and if an authorized face is detected, the door lock is automatically opened. If an unauthorized face is detected, a buzzer is activated to notify security personnel, and an email is sent to an authorized person to alert them of the security breach. The system requires high accuracy in face detection and recognition, and it is essential to ensure the system\'s privacy compliance. The integration of the system with hardware components requires programming skills, and regular maintenance and updates are necessary to ensure the system operates correctly and remains secure.

Introduction

I. INTRODUCTION

Face detection and recognition using a Raspberry Pi is a project that utilizes computer vision and machine learning algorithms to identify human faces and determine if they are authorized to access a specific location or device. The project involves the use of a Raspberry Pi single-board computer, a camera module, and some programming knowledge. The first step in this project is to use the camera module attached to the Raspberry Pi to capture images of a person's face. The images are then processed using face detection algorithms that identify the location and size of the face in the image. Once the face is detected, the system will use face recognition algorithms to compare the detected face to a database of authorized faces. If the detected face matches an authorized face in the database, the door lock will be opened. If the detected face does not match an authorized face, a buzzer will sound, and an email notification will be sent to the authorized person's email address, notifying them of the unauthorized access attempt.

To implement this project, you will need to have a basic understanding of programming languages such as Python, OpenCV, and face recognition libraries. Additionally, you will need a Raspberry Pi board, a camera module, and a door lock or buzzer.

Overall, the face detection and recognition project using Raspberry Pi can be an exciting and useful application of computer vision technology, providing a secure and automated way to control access to a specific location or device.

II. LITERATURE SURVEY

A. Paper: "Raspberry Pi-based face Recognition system for Surveillance Applications"

Authors: Zhao, J., Mathews, S., & Prabhakaran, B.
Abstract: This paper presents a face recognition system based on the Raspberry Pi platform, designed for surveillance applications. The system utilizes the OpenCV library for face detection and recognition tasks. A Raspberry Pi camera module captures live video feed, and the captured frames are processed in real-time for face detection. A trained face recognition model is used to match the detected faces against a database of known individuals. The system also includes a servo motor-controlled door lock mechanism that can be activated upon successful face recognition. The experimental results demonstrate the effectiveness and efficiency of the proposed system for surveillance and access control applications.

B. Paper: "Real-time Drowsiness Detection using Facial Landmarks"

Authors: Davis, E., & Amit, A.
Abstract: This research paper presents a real-time drowsiness detection system using facial landmarks. The system utilizes the dlib library for facial landmark detection and tracking. By analyzing the movements and positions of key facial landmarks, such as the eyes and eyebrows, the system can accurately detect signs of drowsiness. The proposed system, implemented on a Raspberry Pi platform, aims to address the critical issue of drowsy driving by providing timely alerts to the driver. The experimental results demonstrate the effectiveness of the system in detecting drowsiness in real-time scenarios.

C. Paper: "Eigenfaces for recognition"

Authors: Turk, M., & Pentland, A.
Abstract: This seminal paper introduces the concept of Eigenfaces for face recognition. The Eigenfaces method utilizes Principal Component Analysis (PCA) to represent faces in a reduced-dimensional feature space. By projecting the face images onto the Eigenvectors obtained from the training set, a low-dimensional representation of each face is obtained. This representation can be used for face recognition by comparing the Euclidean distance or Mahalanobis distance between the projected face images. The Eigenfaces method, described in this paper, has proven to be computationally efficient and effective for face recognition tasks and has laid the foundation for subsequent advancements in the field.

D. Paper: "Local binary patterns for face description: Application to face recognition"

Authors: Ahonen, T., Hadid, A., & Pietikainen, M.
Abstract: This paper presents an approach called Local Binary Patterns (LBP) for face description and its application to face recognition. LBP encodes local texture information by comparing the intensity values of a pixel with its neighboring pixels. The resulting binary patterns are then used to construct histograms that capture the texture information of facial regions. The proposed method achieves robustness to variations in lighting conditions and facial expressions. The paper provides an in-depth analysis of LBP-based face recognition algorithms and demonstrates their effectiveness on various benchmark datasets. The lightweight nature of LBP makes it suitable for real-time implementation on resource-constrained platforms like the Raspberry Pi.

E. Paper: "Rapid object detection using a boosted cascade of simple features"

Authors: Viola, P., & Jones, M.
Abstract: This paper introduces the Viola-Jones algorithm, which is a widely adopted method for rapid object detection, including face detection. The algorithm utilizes Haar-like features, which are simple rectangular filters that capture local image intensity patterns. These features are used in a cascaded framework, where a series of classifiers are trained to discriminate between faces and non-faces. The cascaded structure allows for efficient processing by quickly rejecting regions that are unlikely to contain faces. The Viola-Jones algorithm achieved high detection rates while maintaining real-time performance. The paper presents the algorithm's formulation, training process, and evaluation results, demonstrating its effectiveness in face detection applications.

III. SYSTEM DESIGN

The project involves the development of a face recognition-based door lock system using components such as a power supply, Raspberry Pi, buzzer, camera module, and door strike mechanism. The Raspberry Pi serves as the main processing unit and control center for the system. The camera module captures live video frames, and computer vision techniques are employed to detect faces in the frames. A face recognition model is trained using a dataset of known faces, and facial features are compared to authenticate recognized faces. If a match is found, a signal is sent to the door strike mechanism to unlock the door. The buzzer provides audible feedback during different system events. The system operates by continuously capturing video frames, detecting faces, performing face recognition, and providing real-time authentication and door lock control. Overall, the system design integrates the components to create a secure door lock system based on face recognition technology.

V. ACKNOWLEDGMENT

We would also like to show our gratitude to, Prof. S. B. Pokharkar (professor, department of Electronics and Telecommunication Engineering, AISSMS Institute of Information Technology, Pune, Maharashtra, India.) for their continuous guidance. We are also grateful to him for his comments on an earlier version of the manuscript, although any errors are our own and should not tarnish the reputations of these esteemed persons.

Conclusion

Face detection and recognition systems using Raspberry Pi for door lock applications have gained significant popularity due to their ability to provide secure access control and enhance overall security. These systems utilize the Raspberry Pi\'s processing power and camera module to capture live video frames. By employing face detection algorithms such as Viola-Jones or dlib, the system detects faces in the captured frames. To ensure accurate recognition, the captured face images undergo feature extraction using techniques like Eigenfaces or LBP. A face recognition model is trained using a dataset of known faces, allowing the system to compare the extracted facial features with the database. If a match is found, an authenticated message is displayed, and a servo motor is triggered to unlock the door. This face detection and recognition system offers a cost-effective and efficient solution for access control, ensuring that only authorized individuals can gain entry. The integration of hardware components, image processing, and face recognition algorithms provides a reliable and secure method for door lock applications.

References

[1] Sandhu, S., Grewal, A., & Singh, G. (2020). A Review on Face Detection and Recognition Techniques. 2020 2nd International Conference on Advances in Science & Technology (ICAST), 1-5. doi: 10.1109/ICAST51156.2020.9280013 [2] Prakash, K. S., Rajendran, R., & Jayakumar, T. (2019). Raspberry Pi based Face Recognition System for Door Access Control. 2019 International Conference on Electrical, Communication, and Computer Technologies (ICECCT), 1-4. doi: 10.1109/ICECCT.2019.8869083 [3] Sarangi, S., & Jena, R. K. (2020). Real-time Face Recognition based Door Lock System using Raspberry Pi. 2020 International Conference on Computer Science, Engineering and Applications (ICCSEA), 1-6. doi: 10.1109/ICCSEA48962.2020.913361 [4] Sayadiyan, Z., & Dashtban, M. (2020). Real-Time Face Detection and Recognition System using Raspberry Pi for Door Access Control. 2020 3rd International Conference on Computer Science and Artificial Intelligence (CSAI), 257-261. doi: 10.1109/CSAI50020.2020.912712 [5] Bae, S., Jung, H., & Son, J. (2019). Real-Time Face Recognition Door Lock System using Raspberry Pi. 2019 International Conference on Platform Technology and Service (PlatCon), 1-5. doi: 10.1109/PlatCon46720.2019.8988936

Copyright

Copyright © 2023 Tejas Tonpe, Tejas Saggam, Nutan Limbare, Prof. Sagar Pokharkar. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.