International Research Journal of Modernization in Engineering Technology and Science
( Peer-Reviewed, Open Access, Fully Refereed International Journal ) Volume:05/Issue:03/March-2023 Impact Factor- 7.868 www.irjmets.com FACIAL AND FINGERPRINT RECOGNITION BASED DOOR LOCK SYSTEM USING RASPBERRY PI AND HOG CLASSIFIER Shivani Narsina*1, Balaji Yashwanth Orre*2, Pradeep Kakkarla*3, J Beatrice Seventline*4, Saiteja Chopparapu*5 *1,2,3Student, Department Of Electrical, Electronics And Communications Engineering, GITAM School Of Technology, GITAM (Deemed To Be University), Visakhapatnam, India – 5300070. *4Professor, Department Of Electrical, Electronics And Communications Engineering, GITAM School Of Technology, GITAM (Deemed To Be University), Visakhapatnam, India – 530007. *5Research Scholar, Department Of Electrical, Electronics And Communications Engineering, GITAM School Of Technology, GITAM (Deemed To Be University), Visakhapatnam, India – 530007. ABSTRACT Facial recognition in security systems plays a key role. This includes the automation of security systems by using bio-metric based door locks and many more applications. In this work, we propose a facial and fingerprint recognition-based door lock system using Raspberry Pi and an Optical fingerprint reader. In this work, HOG classifier is used because of its simplicity and high accuracy. The input images are taken from the sensor which is a camera and it is fed as input to the HOG classifier which is used to classify the images. The Recognizer implemented in Raspberry Pi compares the faces detected in the feed with already existing data of the faces in the trained data set. Similarly, an optical fingerprint reader is integrated into the system to provide 2 step authentication and make the system more accurate. A Mail Alert is received by the owner every time the system recognizes the person as unauthorized personnel in both stages of authentication Keywords: Biometric Authentication, Facial Recognition, Mail Alert, Raspberry Pi. I. INTRODUCTION With the increase in crime rates, home surveillance and security systems have become essential in protecting our homes and loved ones. Home surveillance and security systems can help deter burglars, provide evidence in case of a break-in, and alert authorities in case of an emergency. One of the most important features of modern security systems is facial and biometric recognition technology. These technologies can help identify intruders, monitor who enters and exits your home, and provide an additional layer of security. Facial recognition technology uses artificial intelligence and machine learning algorithms to identify a person based on their facial features. This technology has become increasingly popular in recent years, with many companies using it in their security systems. Facial recognition technology can be used to detect unauthorized access to your home and can be programmed to only allow authorized individuals to enter. This technology can also be used to track visitors to your home and monitor their activities, ensuring that your home remains safe and secure. Biometric recognition technology is another important feature of modern security systems. This technology uses unique biological characteristics, such as fingerprints or iris scans, to identify individuals. Biometric recognition technology can be used to provide access to your home, ensuring that only authorized individuals are allowed to enter. This technology is highly secure and cannot be easily bypassed, making it an effective tool in protecting your home. Facial recognition and biometric-based door lock systems are an excellent example of how technology can be used to enhance home security. These systems use facial recognition and biometric technology to provide access to your home, ensuring that only authorized individuals are allowed to enter. These systems can be programmed to recognize multiple users, making it easy for family members to access your home without the need for keys or passwords. They can also be programmed to deny access to unauthorized individuals, providing an additional layer of security. In conclusion, home surveillance and security systems are essential in protecting your home and loved ones. Facial and biometric recognition technology has become an important feature in modern security systems, providing an additional layer of security and peace of mind. Facial recognition and biometric-based door lock systems are excellent examples of how this technology can be used to enhance home security, providing secure access to your home while keeping intruders out. www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science [2855] e-ISSN: 2582-5208 International Research Journal of Modernization in Engineering Technology and Science ( Peer-Reviewed, Open Access, Fully Refereed International Journal ) Volume:05/Issue:03/March-2023 Impact Factor- 7.868 www.irjmets.com II. LITERATURE SURVEY Facial and biometric recognition systems have become increasingly popular in surveillance systems due to their accuracy and efficiency. This literature survey aims to explore recent research in the field of facial and biometric recognition-based surveillance systems including mail alerts. [1] Singh and Singh (2020) proposed a real-time face recognition system using HOG features and SVM classifier. Their system achieved high accuracy rates in face recognition tasks.[2] Balasubramanian et al. (2020) presented a comprehensive study on fingerprint recognition using convolutional neural networks. Their study showed that deep learning-based approaches can achieve higher accuracy rates compared to traditional methods. [3] Kaur and Vashisht (2020) conducted a comparative study of feature extraction techniques for facial recognition. Their study compared the performance of various feature extraction techniques and concluded that HOG features provide high accuracy rates in facial recognition tasks. [4] Namburu and Kamal (2021) developed a facial recognition system using the HOG feature extraction technique. Their system achieved high accuracy rates and can be used in surveillance systems. [5] Tandel and Patel (2019) conducted a review of deep learning-based approaches for image detection. Their review discussed various deep learning models and their applications in image detection. [6] Mahajan and Mishra (2020) proposed an e-mail alert system using Python. Their system can be used in surveillance systems to notify users of suspicious activities. [7] Sarkar et al. (2021) proposed an efficient face recognition system using HOG and LBP features. Their system achieved high accuracy rates and can be used in surveillance systems. [8] Reddy and Rao (2020) conducted a comparative study of fingerprint recognition systems. Their study compared the performance of various fingerprint recognition systems and concluded that deep learning-based approaches provide higher accuracy rates compared to traditional methods. [9] Singh et al. (2020) conducted a comparative study of face recognition systems using HOG and SVM. Their study concluded that HOG features provide high accuracy rates in face recognition tasks. Overall, the reviewed literature highlights the importance of facial and biometric recognition-based surveillance systems in ensuring security and safety. These systems can be further improved by incorporating advanced feature extraction techniques and deep learning-based approaches. Additionally, the use of e-mail alert systems can enhance the efficiency of surveillance systems by notifying users of suspicious activities in real-time. III. SOFTWARES, LIBRARIES AND COMPONENTS USED 3.1 Open CV: OpenCV (Open Source Computer Vision Library) is a library of programming functions mainly aimed at real-time computer vision. 3.2 DLib: Dlib is a general purpose cross-platform software library written in the programming language C++. Its design is heavily influenced by ideas from design by contract and component-based software engineering 3.3 face_recognition: face_recognition is a open source library that recognizes and manipulate faces from command line or from python using dlib’s state-of-art face recognition. This library provides functions for facial landmark identification, feature encoding and comparison of encodings. It is developed by Adam Geitey. The library is available as a repository in github which can be cloned by command in command line or Git Bash 3.4 numpy: NumPy is a library for the Python programming language, adding support for large, multi – dimensional arrays and matrices, along with a large collection of high – level mathematical functions to operate on these arrays. 3.5 smtplib: The smtplib module defines an SMTP client session object that can be used to send mail to any internet machine with an SMTP or ESMTP listener daemon. For details of SMTP and ESMTP operation, consult RFC 821 (Simple Mail Transfer Protocol) and RFC 1869 (SMTP Service Extensions). 3.6 pyfingerprint: The PyFingerprint library allows to use ZhianTec ZFM-20, ZFM-60, ZFM-70 and ZFM-100 fingerprint sensors on the Raspberry Pi or other Linux machines. Some other models like R302, R303, R305, R306, R307, R551 and FPM10A also work. 3.7 VNC Viewer: VNC Viewer is used for local computers and mobile devices you want to control from. A device such as a computer, tablet, or smart phone with VNC Viewer software installed can access and take control of a computer in another location. 3.8 Raspberry Pi 3b+: The Raspberry Pi 3B+ is a small, single-board computer that is designed for hobbyists, students, and professionals. It features a 1.4GHz quad-core processor, 1GB of RAM, and onboard wireless www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science [2856] e-ISSN: 2582-5208 International Research Journal of Modernization in Engineering Technology and Science ( Peer-Reviewed, Open Access, Fully Refereed International Journal ) Volume:05/Issue:03/March-2023 Impact Factor- 7.868 www.irjmets.com connectivity (including both Wi-Fi and Bluetooth). The device also includes four USB ports, an HDMI port for connecting to a display, and a microSD card slot for storage. The Raspberry Pi 3B+ is a versatile and affordable device that can be used for a wide range of projects, including media centers, game consoles, home automation systems, and more.
Fig. 1: Raspberry Pi 3b+
3.9 R307 Optical Fingerprint Sensor: This is the R307 Optical Fingerprint Reader Sensor Module. R307 fingerprint module is a fingerprint sensor with a TTL UART interface for direct connections to microcontroller UART or to PC through MAX232 / USB-Serial adapter. The user can store the fingerprint data in the module and can configure it in 1:1 or 1: N mode for identifying the person.
Fig. 2: R307 Optical Fingerprint Sensor
3.10 Servo Motor and Buzzer: Servo motors are electronic devices and rotary or linear actuators that rotate and push parts of a machine with precision and a Piezo Buzzer is a type of electronic device that's used to produce a tone, alarm or sound. It's lightweight with a simple construction, and it's typically a low-cost product.
Fig. 3: Servo Motor and Buzzer
3.11 PL2303 USB to TTL Converter: The PL2303 USB to TTL(Serial) Converter is a small electronic device that allows for communication between devices with TTL-level serial interfaces and a computer with a USB port. It is commonly used for programming microcontrollers, communicating with embedded systems, and other similar applications that require serial communication.
Fig. 4: PL2303 USB to TTL Converter
3.12 iBall Face2Face C8.0 Webcam: The iBall Face2Face C8.0 Webcam is a high-definition webcam designed for video conferencing and online communication. It features a 8-megapixel camera, built-in microphone, and easy-to-use plug-and-play setup. The webcam also includes facial recognition technology for added security and convenience. www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science [2857] e-ISSN: 2582-5208 International Research Journal of Modernization in Engineering Technology and Science ( Peer-Reviewed, Open Access, Fully Refereed International Journal ) Volume:05/Issue:03/March-2023 Impact Factor- 7.868 www.irjmets.com
Fig. 5: iBall Face2Face C8.0 Webcam
IV. HOG CLASSIFIER The Histogram of Oriented Gradients (HOG) classifier is a popular feature descriptor used in computer vision and image processing applications. It is often used for object detection and recognition tasks, such as detecting faces in images. The HOG algorithm works by first dividing the image into small, overlapping rectangular regions called cells. Within each cell, the gradient orientation and magnitude of the pixel values are calculated. The gradient orientation is the direction of the steepest increase in pixel intensity, while the gradient magnitude is the magnitude of that increase. Next, the cells are grouped into larger rectangular regions called blocks. Each block contains a certain number of cells, and the gradient information from each cell is used to construct a histogram of gradient orientations for that block. This histogram represents the dominant directions of the gradients within the block. Finally, the HOG classifier uses the histograms of the gradient orientations as features to train a machine learning model, such as a Support Vector Machine (SVM), to distinguish between different objects or classes. The model is trained on a set of labeled images and learns to recognize the patterns of the histograms that correspond to the object or class of interest. During object detection, the HOG algorithm slides a window across the image and computes the HOG features for each window. The classifier then evaluates each window using the machine learning model to determine whether it contains the object of interest. If the window is classified as containing the object, the location and size of the window are used to localize the object within the image. Overall, the HOG classifier is a powerful tool for object detection and recognition tasks, providing robust and accurate results in a variety of applications. V. BLOCK DIAGRAM
Fig. 6: Block Diagram
www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science
[2858] e-ISSN: 2582-5208 International Research Journal of Modernization in Engineering Technology and Science ( Peer-Reviewed, Open Access, Fully Refereed International Journal ) Volume:05/Issue:03/March-2023 Impact Factor- 7.868 www.irjmets.com VI. HARDWARE SETUP
Fig. 7: Hardware Setup Fig 8: Prototype for Automatic Door Lock
VII. RESULTS
Fig. 9: Owners recognized Fig. 10: Unknown person recognized as Intruder
Fig. 11: Mail Alerts
VIII. CONCLUSION The proposed system has a facial recognition and fingerprint-based door lock system that uses Raspberry Pi and an Optical fingerprint reader for biometric authentication. The system uses the HOG classifier algorithm for facial recognition because of its simplicity and high accuracy. The input images are captured by a camera and fed to the HOG classifier to classify the images. The recognizer in Raspberry Pi compares the detected faces with the already existing data of the faces in the trained data set to grant access to authorized individuals. In addition, an optical fingerprint reader is integrated into the system to provide a 2-step authentication process, increasing the accuracy of the system. Whenever a person enters the system, a mail alert is sent to the owner to keep track of the entry. Overall, this system provides a highly secure and accurate method for access control. ACKNOWLEDGEMENTS We thank Prof. J.B. Seventline, HoD, Department of Electrical, Electronics and Communication Engineering, GITAM School of Technology, GITAM (Deemed to be University), Visakhapatnam, India for her immense support and excellent guidance throughout the process.
www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science
[2859] e-ISSN: 2582-5208 International Research Journal of Modernization in Engineering Technology and Science ( Peer-Reviewed, Open Access, Fully Refereed International Journal ) Volume:05/Issue:03/March-2023 Impact Factor- 7.868 www.irjmets.com Our sincere thanks to SaiTeja Chopparapu, Research Scholar, Department of Electrical, Electronics and Communication Engineering, GITAM School of Technology, GITAM (Deemed to be University), Visakhapatnam, India for his timely guidance and support over the period of completion of the project IX. REFERENCES [1] Singh, G., & Singh, R. (2020). “Real-time face recognition using HOG features and SVM classifier. In 2020 International Conference on Computer Communication and Informatics“ (ICCCI) (pp. 1–6). IEEE. [2] Balasubramanian, M., Arun, P., & Ramesh, B. (2020). “A comprehensive study on fingerprint recognition using convolutional neural networks. Journal of Ambient Intelligence and Humanized Computing”, 11(9), 3667–3679. [3] Kaur, G., & Vashisht, S. (2020). “A comparative study of feature extraction techniques for facial recognition”. In 2020 11th International Conference on Computing, Communication and Networking Technologies (ICCCNT) (pp. 1–6). IEEE. [4] Namburu, N. V., & Kamal, T. (2021). “Development of a facial recognition system using the HOG feature extraction technique”. In 2021 3rd International Conference on Intelligent Computing, Instrumentation and Control Technologies (ICICICT) (pp. 188–193). IEEE. [5] Tandel, D. D., & Patel, N. B. (2019). “A review of deep learning based approaches for image detection”. In 2019 5th International Conference on Computing, Communication, Control and Automation (ICCUBEA) (pp. 1–6). IEEE. [6] Mahajan, A., & Mishra, M. (2020). “E-mail alert system using Python”. In 2020 2nd International Conference on Advances in Computing, Communication, Control and Networking (ICACCCN) (pp. 280– 284). IEEE. [7] Sarkar, S., Bhattacharyya, D., & Das, R. (2021). “An efficient face recognition system using HOG and LBP features”. In 2021 International Conference on Electronics, Computing and Communication Technologies (CONECCT) (pp. 1–6). IEEE. [8] Reddy, S. K., & Rao, P. M. (2020). “A comparative study of fingerprint recognition systems”. In 2020 International Conference on Emerging Trends in Information Technology and Engineering (ic-ETITE) (pp. 1–5). IEEE. [9] Singh, S., Kumar, A., & Singh, G. (2020). “Face recognition system using HOG and SVM”: A comparative study. In 2020 International Conference on Computer Science, Engineering and Applications (ICCSEA) (pp. 1–6). IEEE. [10] A Geitgey “Machine Learning is Fun Part 4 – Modern Face Recognition with Deep Learning” [11] SaiTeja Chopparapu, Dr. Beatrice Seventline J. “Object Detection using MATLAB and Python”. International Journal of Electrical Engineering and Technology, 11(6),2020, pp. 101-108 [12] SaiTeja Chopparapu and Dr. Beatrice Seventline J. “GUI for Object Detection using Voila Method in MATLAB”. International Journal of Electrical Engineering and Technology, 11(4),2020, pp. 169-174 [13] Harihara Santosh Dadi, Gopala Krishna Mohan Pillutla – “Improved Face Recognition Rate Using HOG Features and SVM Classifier”
www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science