Smart CCTV Surveillance using Telegram Bot

Main Article Content

Nitesh, Gaurisha, Komal Malsa

Abstract

In a world rapidly shifting toward smart living, relying solely on traditional CCTV surveillance no longer meets the demands of dynamic, real-time security. This research introduces an innovative smart CCTV surveillance system, powered by facial recognition and enhanced through Telegram Bot integration. The primary goal is to deliver an interactive, user involved monitoring solution capable of recognizing familiar faces, identifying intruders, and updating its own database based on user feedback all in real time.


The core mechanism classifies individuals captured on camera into “known” and “unknown” categories. When the system spots an unknown person, it instantly sends a notification along with an image to the authorized user through Telegram. From there, the user has the option to tag the face as known, prompting the system to automatically create a new folder for that individual and begin storing their images under their assigned identity. This unique feature ensures the database grows over time with minimal user effort and no need for technical intervention.


We’ve used pre-trained, high-accuracy facial recognition models to improve detection even under different lighting or environmental conditions. Thanks to Telegram’s secure bot infrastructure, communication between the system and the user is fast, reliable, and private. Access is restricted to verified users, preserving data integrity and control.


Extensive testing across various real-world conditions confirmed the system’s reliability and responsiveness. Its hands-on interactivity allows even non-technical users to contribute to system learning, making it practical for a wide range of uses from private residences and small businesses to schools and office buildings. In this by combining accessible technology with modern AI-based recognition, this project sets  the way for next generation surveillance tools that are intelligent, scalable, and user-focused.

Article Details

Section
Articles