MUMBAI, India, Jan. 9 -- Intellectual Property India has published a patent application (202541134126 A) filed by Karpagam Academy Of Higher Education; Karpagam Institute Of Technology; Dr. V Vadivu; and Mohammed Ashick M, Coimbatore, Tamil Nadu, on Dec. 31, 2025, for 'intruder's detection system using machine learning.'
Inventor(s) include Dr. V Vadivu; and Mohammed Ashick M.
The application for the patent was published on Jan. 9, under issue no. 02/2026.
According to the abstract released by the Intellectual Property India: "Intruder detection has become an essential component of modern security infrastructures due to the increasing frequency of unauthorized access, physical breaches, and cyber-physical threats across residential, industrial, and organizational environments. Traditional security systems rely heavily on predefined rules, threshold-based triggers, and manual supervision, which significantly limit their ability to detect sophisticated or evolving intrusion patterns. To address these challenges, this research proposes an Intelligent Intruder Detection System (IDS) leveraging advanced Machine Learning (ML) techniques for real-time, automated threat identification. The proposed system integrates multimodal data sources, including surveillance camera feeds, motion sensor outputs, and network activity logs, enabling comprehensive environmental monitoring. A Convolutional Neural Network (CNN)-based visual recognition module processes image and video data to detect human presence, abnormal motion patterns, and unauthorized entry. Simultaneously, an anomaly detection framework incorporating Isolation Forest and Autoencoder models evaluates sensor and network data to identify deviations from established behavioral norms. The fusion of supervised and unsupervised learning enables robust detection of both known and previously unseen intrusion attempts. The system employs a hybrid decision-making engine that combines the outputs of visual and anomaly detection modules to generate accurate and reliable intrusion alerts. A continuous learning mechanism further enhances adaptability by updating model parameters with new data and environmental variations. Real-time notifications are triggered through alarms or digital alerts to ensure rapid response. Performance evaluation is conducted using metrics such as accuracy, precision, recall, and F1-score to validate system reliability and minimize false positives. The architecture is designed to be scalable and seamlessly deployable across diverse applications, including smart homes, industrial sites, and enterprise facilities. By integrating intelligent analytics with multimodal data processing, the proposed IDS offers a highly efficient, adaptable, and resilient solution for next-generation security systems."
Disclaimer: Curated by HT Syndication.
