Network, VANET,Data Mining, Machine Learning and Programming
8
Scopus Publications
Scopus Publications
Enhancing bidirectional gated recurrent unit with activation mechanism for anomaly classification for network security N. Senthil MADASAMY, A. Noble Mary JULIET, P. Boopathi RAJAN Sigma Journal of Engineering and Natural Sciences, 2026 In a digital world, cyber-attacks are increasingly common, raising concerns that existing anomaly detection models might not effectively handle intricate threat scenarios.As the demands for network systems grow.Historically, the update and reset gates of the Gated Recurrent Unit (GRU) designed to controls flow in input data across time steps have faced challenges in identifying anomalies in security monitoring, traffic log analyses, and packet flow assessments.To address anomalies, reduce time expenditure, and improve network security, the proposed research utilizes a Deep Learning (DL) technique named Structured Activation Module Loop Framework Unit and an efficient activation module unit, which integrates a Bidirectional Gated Recurrent Unit (Bi-GRU) model that includes update and reset gates for controlling information flow on the basis of the classification context of the input data.The suggested structured activated loop framework monitors error data straight in the update gate without demanding the reset gate, enabling several checks in a loop format.The activation module unit precisely divides the class data to predict the appropriate output characteristics for resolving missing values.This utilizes network intrusion datasets (UNSW-NB15) and Neural Simulation Language-Knowledge Discovery in Databases (NSL-KDD), both commonly used for (NID) Network Intrusion Detection systems, along with pre-processing data and assessments for data splitting through training the model and testing procedures.Similarly, the proposed performance is assessed using different metrics such as F1-meaures, recall, value of precision, accuracy, with overall accuracy to evaluate the efficiency of the suggested deep learning study.The results obtained precisions of 0.99, 0.97, 0.97, and 0.97 in NSL-KDD, 0.98, 0.99, 0.98, and 0.98 in UNSW-NB 15.Still, the assessment of recommended models demonstrates the efficacy of the research.The present research attempts for find and improve creation of detection of anomaly models for network and cyber security regarding hackers and malicious attacks.
SAFE (Smart Alert Framework for Elephant) - Ai Based Human-Wildlife Conflict Prevention Using Yolo: A Review and Analysis F. Sofiah, D. Sivaganesan, N. Suba Rani, A.Noble Mary Juliet, N. Senthil Madasamy, J. Bhavithra 2025 IEEE 1st International Conference on Innovations in Engineering and Next Generation Technologies for Sustainability Icinvents 2025, 2025 Human-Elephant conflict is a critical issue in wildlife conservation and rural agriculture, particularly in regions with dense elephant populations. Conventional monitoring methods such as manual patrolling and camera traps lack real-time capabilities and often fail to prevent damage and loss. This paper presents a comprehensive review of existing AI-based wildlife monitoring techniques, with a focus on YOLO (You Only Look Once) models, edge computing, and IoT-based alert systems. The study compares 20 recent research papers highlighting advances in real-time animal detection, emotion analysis, and low-cost scalable solutions for rural areas. Our analysis identifies gaps in behavioral prediction and edge-device deployment, leading to a proposal for a Smart Alert Framework for Elephants (SAFE) that ensures early detection, aggression prediction, and realtime alerts to farmers and forest officers
AI-Powered Real-Time Runway Safety: UAV-Based Video Analysis with ICSO-Enhanced Deep Learning V. Thamilarasi, R. Hema, A. Noble Mary Juliet, Adlin Sheeba, Gauri Ghule, A. Raja International Journal of Computational and Experimental Science and Engineering, 2024 In the aviation sector, ensuring safe landings while prioritizing the safety of runways is crucial to prevent accidents and incidents during the landing phase of flights. However, many studies analyzing unsafe events, such as runway cracks or inadequate friction, often fail to quantify their impacts on flight safety during landing. In airport pavement management systems (APMS), the condition of the runway surface is a critical factor in ensuring the operational safety of aircraft during take-off and landing. Therefore, it is essential to provide pilots with reports on runway conditions, including measurements of surface performance, to support informed decision-making. To tackle these challenges, we propose a real-time automatic monitoring system for runway safety utilizing video analysis. Specifically, we employ a time-series analysis approach using the improved chameleon swarm optimization (ICSO) algorithm to mine runway surface characteristics from real-time video data captured by unmanned aerial vehicles (UAVs). Subsequently, we introduce the fuzzy reinforced polynomial neural network (FR-PNN) to detect risks in runway surface characteristics, enabling automatic monitoring to enhance the safety of aircraft landings. Finally, the effectiveness of the proposed system is validated using real-time videos obtained from Bechyne military airport, located in Bohemia. This system aims to improve runway safety by providing timely and accurate assessments of runway conditions, thereby facilitating safer landings for aircraft.
An Image Encryption and Decryption And Comparison With Text - AES Algorithm International Journal of Scientific and Technology Research, 2019
Optimized searching of video based on speech and video text content G. Vigneshwari, A. Noble Mary Juliet Proceedings of the IEEE International Conference on Soft Computing and Network Security Icsns 2015, 2015 Digitalized video has become an established storage and exchange medium due to the fast development in recording technology. E-lecturing has become added professional popular. The number of lecture video data on the web is growing quickly. Therefore, an additional skillful method for video retrieval within huge lecture video archives is immediately desired. This paper comes close to regular video indexing and video stalk in large lecture videos collection. We apply Automatic video segmentation and key-frame detection to present an image instruction for the video content navigation. In this process, video will be retrieved based on content based video search for the input of speech and video text content. Histogram of gradient and Support Vector Machine is used for the feature extraction from the video and to get the accurate result respectively. This will collect required features from lecture video database as labeled with speech recognition and also retrieve lecture video from database as better than existing algorithm. Hence the proposed system will be more efficient for retrieving the videos and also improves the recognition rate.
Neighbor based rebroadcasting for vanet European Journal of Scientific Research, 2012
