Keys, access cards and PIN codes can be misplaced, duplicated or handed to others, and an ordinary lock offers the owner no means of judging a person who has not been registered. This paper proposes a software-simulated smart door lock that admits registered people through face recognition and forwards every unregistered visitor to the owner for a decision. A webcam provides live frames to a Haar Cascade detector. Each face found is normalised and compared with a stored set of enrolled people by the Local Binary Patterns Histograms (LBPH) recognizer in OpenCV, and a distance threshold separates familiar faces from strangers. An enrolled person opens a virtual door at once. For a stranger, the program sends the captured image to the owner’s Telegram bot together with Allow and Deny buttons, and the door opens only after Allow is pressed; Deny, silence or any fault leaves it locked. All events are recorded in a time-stamped log. The prototype runs in Python on an ordinary laptop, so no Raspberry Pi board or lock hardware is required during testing, while the modular code can later drive a relay. The paper explains the architecture, algorithms, parameter choices, security considerations, expected strengths and weaknesses, and a structured test plan. No measured accuracy is reported, because the prototype has not yet been evaluated.
Introduction
The proposed project focuses on developing a Face Recognition-Based Smart Door Lock System with Remote Access Approval Using Telegram. The system aims to improve traditional security methods by using facial recognition to identify authorized individuals and allowing the owner to remotely approve or deny entry to unknown visitors.
Traditional security systems, such as mechanical keys, passwords, and access cards, have several limitations because they can be lost, copied, shared, or stolen. Biometric authentication, particularly face recognition, provides a more personalized approach because it identifies individuals based on their facial features.
Main Objectives
The primary objectives of the project are:
Face Detection: Detect human faces through a webcam in real time.
Face Recognition: Identify registered users and automatically unlock the virtual door for authorized individuals.
Remote Approval: Send a photograph of an unknown visitor to the owner's Telegram account with Allow and Deny options.
Secure Access: Keep the door locked unless an authorized person is recognized or the owner explicitly approves entry.
Activity Logging: Record every access attempt, including the time, visitor identity, and access outcome.
Modular Design: Develop an independent software system that can later be connected to a physical door lock using a relay.
Related Technologies and Research
The project uses existing face detection and recognition techniques:
Viola-Jones Algorithm: Used for detecting faces efficiently in webcam video.
Eigenfaces: A face recognition technique based on principal component analysis (PCA).
Fisherfaces: Improves recognition under varying lighting and facial expressions.
Local Binary Patterns Histogram (LBPH): Used in the proposed system to recognize faces based on local facial texture features.
Deep Learning-Based Recognition: Techniques such as FaceNet provide advanced recognition accuracy but require greater computational resources.
The Telegram Bot API is used to send visitor photographs and provide remote access approval through interactive buttons.
Working Principle
The system follows a simple workflow:
The webcam captures the visitor's image.
The face detection module identifies the face.
The recognition module compares it with registered users.
If the person is recognized, the virtual door unlocks automatically.
If the person is unknown, their photograph is sent to the owner's Telegram account.
The owner selects Allow or Deny.
The system processes the decision and records the access attempt.
If an error occurs or approval is not received within the specified time, the door remains locked.
Security and Evaluation
The proposed system includes security measures such as owner-only approval, unique request identification, time-limited approval, and a fail-safe locking mechanism.
Its evaluation will consider face recognition performance, response time, system reliability, approval handling, and security. The project also plans to examine possible threats and their countermeasures.
Conclusion
This paper proposed a smart door lock in which enrolled people are admitted by face recognition and unknown visitors are referred to the owner through a Telegram bot with Allow and Deny buttons. Haar cascade detection, LBPH recognition and a distance threshold keep the system light enough for a laptop, while fail-safe rules and the owner’s final decision limit the consequence of recognition errors. The modular design allows the virtual door to be exchanged for a physical one.
The usefulness of the design depends on good enrolment, a carefully chosen threshold, stable network access and sensible handling of repeated detections. The prototype should be evaluated according to Section X, and real measurements should be reported before any claim is made about accuracy or response time.
References
[1] P. Viola and M. Jones, “Rapid object detection using a boosted cascade of simple features,” in Proc. IEEE Conf. Comput. Vis. Pattern Recognit. (CVPR), Kauai, HI, USA, 2001, pp. I-511–I-518.
[2] R. Lienhart and J. Maydt, “An extended set of Haar-like features for rapid object detection,” in Proc. IEEE Int. Conf. Image Process. (ICIP), Rochester, NY, USA, 2002, pp. I-900–I-903.
[3] G. Bradski, “The OpenCV library,” Dr. Dobb’s J. Softw. Tools, 2000.
[4] M. Turk and A. Pentland, “Eigenfaces for recognition,” J. Cogn. Neurosci., vol. 3, no. 1, pp. 71–86, 1991.
[5] M. A. Turk and A. P. Pentland, “Face recognition using eigenfaces,” in Proc. IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit. (CVPR), Maui, HI, USA, 1991, pp. 586–591.
[6] T. Ojala, M. Pietikäinen, and T. Mäenpää, “Multiresolution gray-scale and rotation invariant texture classification with local binary patterns,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 24, no. 7, pp. 971–987, Jul. 2002.
[7] T. Ahonen, A. Hadid, and M. Pietikäinen, “Face description with local binary patterns: Application to face recognition,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 28, no. 12, pp. 2037–2041, Dec. 2006.
[8] T. Ahonen, A. Hadid, and M. Pietikäinen, “Face recognition with local binary patterns,” in Proc. Eur. Conf. Comput. Vis. (ECCV), Prague, Czech Republic, 2004, pp. 469–481.
[9] F. Schroff, D. Kalenichenko, and J. Philbin, “FaceNet: A unified embedding for face recognition and clustering,” in Proc. IEEE Conf. Comput. Vis. Pattern Recognit. (CVPR), Boston, MA, USA, 2015, pp. 815–823.
[10] Telegram, “Telegram Bot API.” [Online]. Available: https://core.telegram.org/bots/api. Accessed: Nov. 3, 2026.
[11] A. K. Jain, A. Ross, and S. Prabhakar, “An introduction to biometric recognition,” IEEE Trans. Circuits Syst. Video Technol., vol. 14, no. 1, pp. 4–20, Jan. 2004.
[12] P. N. Belhumeur, J. P. Hespanha, and D. J. Kriegman, “Eigenfaces vs. Fisherfaces: Recognition using class specific linear projection,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 19, no. 7, pp. 711–720, Jul. 1997.
[13] European Parliament and Council, “Regulation (EU) 2016/679 (General Data Protection Regulation),” Official Journal of the European Union, L 119, 2016.
[14] Government of India, “The Digital Personal Data Protection Act, 2023,” Act No. 22 of 2023, The Gazette of India, Aug. 2023.
[15] ISO/IEC 30107-3:2017, “Information technology — Biometric presentation attack detection — Part 3: Testing and reporting,” 2017.