Automatic Image Tagging and Captioning Using Transformer-Based Vision-Language Models Manjushri Joshi, Arpit Agrawal, Muthukumar T, Syed Fahar Al, Rajesh Raikwar, Vashisht Singh 2025 International Conference on Emerging Trends in Networks and Computer Communications Etncc 2025 Proceedings, 2025 The rapid expansion of visual data in sectors like healthcare, e-commerce, and social media increases the demand for efficient photo tagging and labelling systems. Many of the automatic photo tagging and labelling techniques in use today struggle with the complex relationships between visual content and natural language, which reduces their accuracy and scalability. Recent advances in transformerbased models, particularly Vision-Language Transformers (ViLT), have greatly simplified the interaction between images and textual claims. These models simultaneously handle visual and textual input using transformers. This improves feature extraction and semantic matching. This paper investigates the automated tagging and description of pictures using transformer-based vision-language models. Our proposed approach generates tags and descriptions for pictures that fit in their present context by combining modern vision transformers with language models. The system is made up of two main parts: a vision encoder that takes in a picture and pulls out visual features; and a text decoder that uses the extracted features to make useful subtitles or tags. We also present a multi-modal training approach that lets the model learn from both written and visual data at the same time. This makes it better at many real-world tasks. We did a lot of tests on standard datasets to show that our suggested model is much better than current ones at making subtitles and tags that are accurate, fluent, and relevant. The results show that transformer-based vision-language models can be used to automatically understand images and create material for a wide range of purposes.
An Enhanced Deep Learning Framework for Effective Arrhythmia Classification via the Electrocardiogram Signal Vinoth Murugan, Damodar Panigrahy, D Dinesh Kumar, Muthukumar T, A. Senthilkumar, Elamurugan P Proceedings of 6th International Conference on Iot Based Control Networks and Intelligent Systems Icicnis 2025, 2025 Heart disease is a major cause of the increasing death rate around the globe. The electrocardiogram (ECG) is used to provide information about various heart diseases. This study presents the classification of heart disease, such as arrhythmias, using a novel framework that is based on robust empirical mode decomposition (REMD) with a deep convolutional neural network (CNN) + bidirectional long short-term memory (BiLSTM) + convolutional block attention module (CBAM). The low-frequency drift is contaminated with the raw ECG signal due to the patient’s movements and improper electrode selection. The REMD technique is used to eliminate low-frequency drift. Deep CNN is used to identify the spatial features, and BiLSTM captures the temporal information present in the preprocessed ECG signal. The CBAM layer focuses on the most important features through channel and spatial attention. The developed framework is assessed using the MIT-BIH arrhythmia database. The developed framework achieves an accuracy of 99.28% and an F1 score of 99.27% compared to existing techniques. The developed framework is suitable for automatic classification of arrhythmia in clinical practice.
IoT based Smart U-Turn Vehicle Accident Prevention System G Pradeepkumar, G Praveen Santhoshkumar, C Rohith Bhat, M Jeyalakshmi, T Muthukumar, Neelam Sanjeev Kumar 2nd International Conference on Sustainable Computing and Data Communication Systems Icscds 2023 Proceedings, 2023 Unintentional deaths occur at a very high rate in developing countries. Curved roads have significantly more fatalities than straight roads. This occurs mainly on U-turns, hairnin turns, and narrow mountain roads. Drivers in this position cannot see the vehicle approaching from the opposite direction. As a result, thousands of people are killed in car accidents every year. The best way to avoid further accidents is to alert the car driver approaching from the side. Place the ultrasonic range detection sensor on one side of the road before the bend and the light indicator system on the opposite side after the bend. When a vehicle approaches from afar, an ultrasonic sensor on one side of the road sends a signal to the other side of the road via a light system. In response to a warning, the driver may stop the car until the other vehicle has passed. A buzzer will also be used to warn the driver of the car that is approaching.
Design and Development of an Intelligent Energy Management System for a Smart Grid to Enhance the Power Quality Nisha Vasudevan, Vasudevan Venkatraman, A. Ramkumar, T. Muthukumar, A. Sheela, M. Vetrivel, R. J. Vijaya Saraswathi, F. T. Josh Energy Engineering Journal of the Association of Energy Engineering, 2023 MigroGrid (MG) has emerged to resolve the growing demand for energy. But because of its inconsistent output, it can result in various power quality (PQ) issues. PQ is a problem that is becoming more and more important for the reliability of power systems that use renewable energy sources. Similarly, the employment of nonlinear loads will introduce harmonics into the system and, as a result, cause distortions in the current and voltage waveforms as well as low power quality issues in the supply system. Thus, this research focuses on power quality enhancement in the MG using hybrid shunt filters. However, the performance of the filter mainly depends upon the design, and stability of the controller. The efficiency of the proposed filter is enhanced by incorporating an enhanced adaptive fuzzy neural network (AFNN) controller. The performance of the proposed topology is examined in a MATLAB/Simulink environment, and experimental findings are provided to validate the effectiveness of this approach. Further, the results of the proposed controller are compared with Adaptive Fuzzy Back-Stepping (AFBS) and Adaptive Fuzzy Sliding (AFS) to prove its superiority over power quality improvement in MG. From the analysis, it can be observed that the proposed system reduces the total harmonic distortion by about 1.8%, which is less than the acceptable limit standard.
Sensor fusion for IoT based intelligent agriculture system 12th International Conference on Advances in Computing Control and Telecommunication Technologies Act 2021, 2021
RECENT SCHOLAR PUBLICATIONS
An IoT based forest fire detection system using integration of cat swarm with LSTM model R Mahaveerakannan, C Anitha, AK Thomas, S Rajan, T Muthukumar, ... Computer Communications 211, 37-45 , 2023 2023 Citations: 100
Sensor Fusion for IoT based Intelligent Agriculture System SN T.Muthukumar, S.PraneshRaj, A.Raman, G.Thirunavukarasu Grenze International Journal of Engineering and Technology (GIJET) 7, 843-847 , 2021 2021
Security Monitoring in Coal Mining using Wireless Underground Sensor Network DKK T.Muthukumar, S.Arun Grenze International Journal of Engineering and Technology , 2019 2019
Efficiency Improvement of Partially Shaded PV System SS J.Vinoth, T.Muthukumar, M.Muruganandam International Journal of Innovative Research in Science, Engineering and … , 2015 2015 Citations: 2
Implementation of Distributed Maximum Power Point Tracking for Partially Shaded PV Systems DMM J.Vinoth, T.Muthukumar International Journal of Advanced Research in Electrical, Electronics and … , 2015 2015 Citations: 6
Offline Sensorless Control for BLDC-PM Motor Using Finite Element Method RS P.M. Manikandan, G. Praveen Santhoshkumar, T. Muthukumar International Journal of Advanced and Innovative Research 3 (11), 279-282 , 2014 2014
Proportional Integral and Derivative Controller for BLDC Motor SS T.Saarulatha, V.Yaknapriya, T.Muthukumar International Journal of Research in Advent Technology 2 (12), 1-4 , 2014 2014 Citations: 2
A Novel Approach on MIP Technique for Risk Constrained Co-ordinated Scheduling of GENCO SS P.Sankar, T.Muthukumar International Journal of Engineering Research & Technology (IJERT) 2 (Issue … , 2013 2013
MOST CITED SCHOLAR PUBLICATIONS
An IoT based forest fire detection system using integration of cat swarm with LSTM model R Mahaveerakannan, C Anitha, AK Thomas, S Rajan, T Muthukumar, ... Computer Communications 211, 37-45 , 2023 2023 Citations: 100
Implementation of Distributed Maximum Power Point Tracking for Partially Shaded PV Systems DMM J.Vinoth, T.Muthukumar International Journal of Advanced Research in Electrical, Electronics and … , 2015 2015 Citations: 6
Efficiency Improvement of Partially Shaded PV System SS J.Vinoth, T.Muthukumar, M.Muruganandam International Journal of Innovative Research in Science, Engineering and … , 2015 2015 Citations: 2
Proportional Integral and Derivative Controller for BLDC Motor SS T.Saarulatha, V.Yaknapriya, T.Muthukumar International Journal of Research in Advent Technology 2 (12), 1-4 , 2014 2014 Citations: 2
Sensor Fusion for IoT based Intelligent Agriculture System SN T.Muthukumar, S.PraneshRaj, A.Raman, G.Thirunavukarasu Grenze International Journal of Engineering and Technology (GIJET) 7, 843-847 , 2021 2021
Security Monitoring in Coal Mining using Wireless Underground Sensor Network DKK T.Muthukumar, S.Arun Grenze International Journal of Engineering and Technology , 2019 2019
Offline Sensorless Control for BLDC-PM Motor Using Finite Element Method RS P.M. Manikandan, G. Praveen Santhoshkumar, T. Muthukumar International Journal of Advanced and Innovative Research 3 (11), 279-282 , 2014 2014
A Novel Approach on MIP Technique for Risk Constrained Co-ordinated Scheduling of GENCO SS P.Sankar, T.Muthukumar International Journal of Engineering Research & Technology (IJERT) 2 (Issue … , 2013 2013