MVM. D. Vimalapriya

M. D. Vimalapriya

*This is an algorithmically generated author record based on publicly available information.

M. D. Vimalapriya has a total of 11 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    TOPOGUARD-EVOCHAIN: Policy-Aware Federated Learning with Topological Representations and Blockchain-Based Model Lineage

    by Jagan Appavoo, Sivasankar Manikkam, Vimalapriya, Haritha Vasudevan, Praveenkumar G D, Sangita Gautam Lade

    Modern decentralized web systems have grown so complex that managing data across them safely has become a serious challenge. This work presents TopoGuard-EvoChain, a new federated learning framework built to support secure and transparent data governance within JavaScript-based distributed environments. The approach blends three key ideas. First, it uses TopoSketch to capture behavioral patterns as compact topological summaries. Second, it embeds governance constraints directly into local training by means of Differentiable Policy Automata. Third, it employs an Evolutionary Secure Aggregator, which relies on blockchain verification to fuse model updates and maintain an auditable lineage. Experiments were conducted on the CICIDS2017 dataset together with synthetic governance metadata across a five-node hybrid federated setup. The framework reached 98.62 percent accuracy, 98.18 percent F1-score, and a 97.56 percent governance-compliance rate, outperforming FedAvg, FedProx, and BlockFL-GovNet. The proposed framework reduced the reported communication cost by 32.6%, and the blockchain layer achieved 98.6 percent verification success with reduced ledger growth. The model also showed 95.6 percent explanation fidelity and 96.4 percent audit-trail reconstruction accuracy, supporting its reliability in traceable decision making. Overall, the study shows that combining topological learning with blockchain-based tracking can significantly strengthen predictive performance and accountability in federated systems.

  • Autonomous face mask detection using single shot multibox detector, and ResNet-50 with identity retrieval through face matching using deep siamese neural network

    by S. Vignesh Baalaji, S. Sandhya, S. A. Sajidha, V. M. Nisha, M. D. Vimalapriya, Amit Kumar Tyagi

    The COVID-19 pandemic poses a global health challenge. The World Health Organization states that face masks are proven to be effective, especially in public areas. Real-time monitoring of face masks is challenging and exhaustive for humans. To reduce human effort and to provide an enforcement mechanism, an autonomous system has been proposed to detect non-masked people and retrieve their identity using computer vision. The proposed method introduces a novel and efficient method that involves fine-tuning the pre-trained ResNet-50 model with a new head layer for classification between masked and non-masked people. The classifier is trained using adaptive momentum optimization algorithm with decaying learning rate and binary cross-entropy loss. Data augmentation and dropout regularization are employed to achieve best convergence. During real-time application of our classifier on videos, a Caffe face detector model based on Single Shot MultiBox Detector is used to extract the face regions of interest from each frame, on which the trained classifier is applied for detecting the non-masked people. The faces of these people are then captured, which is passed on to a deep siamese neural network, based on VGG-Face model for face matching. The captured faces are compared with the reference images from the database, by extracting the features and calculating cosine distance. If the faces match, the details of that person are retrieved from the database and displayed on the web application. The proposed method has secured best results where the trained classifier has achieved 99.74% accuracy, and the identity retrieval model achieved 98.24% accuracy.