Radharani Saravanakumar

@kongunaducollege.ac.in

Assistant Professor Computer Science
Kongunadu Arts and Science College

Radharani Saravanakumar


               Download Compact PDF  
https://researchid.co/radhasaravanan2010

EDUCATION

Ph.D. Computer Science

RESEARCH, TEACHING, or OTHER INTERESTS

Computer Science, Language and Linguistics
6

Scopus Publications

97

Scholar Citations

3

Scholar h-index

2

Scholar i10-index

Scopus Publications

  • CSBAM-MobileNet: a lightweight attention-enhanced deep learning model for student attentiveness classification
    Sajja Radharani, Venkatramaphanikumar Sistla, Venkata Krishna Kishore Kolli
    Engineering Science and Technology an International Journal, 2026
    Emotions significantly influence behaviour and decision-making, making the assessment of student attentiveness a critical factor in optimizing learning outcomes. This paper introduces a vision-based deep learning framework designed to evaluate student attentiveness through facial expression analysis. Leveraging a Customized lightweight MobileNet architecture enhanced with Channel and Spatial Attention Mechanisms (CSBAM), the model efficiently captures subtle facial cues relevant to attentiveness, such as eye closure, gaze direction, and mouth movements. A custom classroom expression dataset was curated for training, enabling the model to achieve a high accuracy of 91.2%. On more challenging subsets exhibiting varied lighting, occlusion, and head pose variations, the model maintained robust performance with 84.2% accuracy. Evaluation across multiple public benchmark datasets further demonstrated the model’s adaptability, achieving accuracies of 92.5% on DAiSEE, 89.6% on RAF-DB, and 87.4% on FER-2013.This framework offers real-time insights into student engagement, providing a foundation for adaptive and personalized teaching strategies. Future research aims to integrate additional behavioural and physiological cues, such as body posture and voice features, to enable a more comprehensive and multi-modal assessment of student attentiveness. The proposed approach holds promise for enhancing educational interventions and fostering active learning environments.
  • HEViT: A Hybrid Efficient Vision Transformer for Student Attentiveness Detection in Classroom Environments
    Sajja Radharani, Venkatramaphanikumar Sistla, Venkata Krishna Kishore Kolli
    Proceedings of the 23rd IEEE International Conference on Computer Applications Icca 2026, 2026
    Accurate assessment of student attentiveness is essential for enhancing learning outcomes, but conventional observational methods are limited by subjectivity and scalability. To address these challenges, we present a deep learning-based framework that automatically quantifies student attention through facial expression analysis. The framework is evaluated on two datasets: the publicly available DAiSEE dataset and a custom Spontaneous Classroom Expressions Dataset, which contains diverse facial expressions representing attentive and inattentive states collected under realistic classroom conditions. We propose HEViT (Hybrid Efficient Vision Transformer), a hybrid architecture that integrates a custom EfficientNet backbone with a Vision Transformer. The EfficientNet module extracts fine-grained local features from facial regions, while the transformer component captures longrange dependencies and global contextual cues. Experimental results show that HEViT significantly outperforms baseline CNNs, standalone EfficientNet, and ViT models. On DAiSEE, HEViT achieves an accuracy of 90%, and on the Spontaneous Classroom Expressions Dataset, it achieves 88%. These results demonstrate that HEViT provides a robust, scalable, and real-time solution for emotion-aware classroom monitoring systems. Future work will explore integration of multimodal cues such as head and body posture to further improve attentiveness detection and generalization across varied classroom environments.
  • Utilizing Transformers for Enhanced Disaster Response in Multimodal Tweet Classification
    Uddagiri Sirisha, Thulasi Bikku, Sajja Radharani, Venkata Nagaraju Thatha, S. Phani Praveen
    International Journal on Engineering Applications, 2025
    This study aims to discover how to use social media more effectively for crisis response and recovery. Information is gathered and disseminated by using social media due to advancements in information and communication technologies. A technique for identifying useful tweets among user's social media posts is introduced. Assuming that the useful tweets can be located, they can be used by emergency personnel to understand the situation better and take appropriate recovery measures. Most prior studies have analyzed textual data or examined the accompanying visuals in tweets. Research shows that text and visuals provide complementary information to one another. A deep learning framework that uses user-generated tweets as input and an accompanying image is proposed. The primary goal of this paper has been to develop a more effective method of multimodal fusion. The proposed system incorporates visual and textual representations based on a transformer concept. In addition to RoBERTa for text, Vision Transformer for images, Bi-LSTM, and an attention mechanism are also used. An additive and multiplicative fusion method is proposed to combine the strengths of both image and text inputs. Seven datasets, including natural calamities like wildfires, hurricanes, earthquakes, and floods, have been used in extensive tests on several network designs. Regarding accuracy, the presented system has been 94% to 98% better than several state-of-the-art methods. The findings have demonstrated that a deep learning classifier can benefit from identifying interactions between numerous related modalities.
  • Efficient deep learning model for recognize artists voice
    International Journal of Advanced Science and Technology, 2019
  • A novel framework for investigation of cloud attacks
    International Journal of Advanced Science and Technology, 2019
  • A novel enhanced ensemble clustering techniques in machine learning and data mining
    Journal of Advanced Research in Dynamical and Control Systems, 2019

RECENT SCHOLAR PUBLICATIONS

  • Implementation of Digital Watermarking with Genetic Algorithm
    DS Radharani
    Sangam 13 (7-8), 153-159 , 2025
    2025.0
  • Study on Medical Image Watermarking
    DTR Dr. S. Radharani, M. Umadevi
    HTL JOURNAL 29 (9), 485 - 495 , 2023
    2023.0
  • Content-Based Watermarking Using MCA
    S Radharani, ML Valarmathi
    Proceedings of International Conference on ICT for Sustainable Development … , 2016
    2016.0
  • Content based Medical Image Watermarking Scheme Based on Arnold Transform with DWT Blending in Block SVD and Expansion for still images using ICA and Random blocks
    DMLV S. Radharani
    International Journal of Applied Engineering Research 10 (1), 1217-1227 , 2015
    2015.0
  • SCANNED DOCUMENT COMPRESSION USING HIGH EFFICIENCY VIDEO CODING (HEVC) STANDARD
    SR B.Nithya
    International Journal of Advances in Computer Science and Technology 3 (11 … , 2014
    2014.0
    Citations: 4
  • Content based Image Watermarking Scheme using Block SVD and Arnold Transform
    DMLV S. Radharani
    International Conference on Electronics and Communications Systems 1 (1 … , 2014
    2014.0
    Citations: 3
  • Content Based Watermarking Techniques using HSV and Fractal Dimension in Transform Domain
    S Radharani, ML Valarmathi
    Australian Journal of Basic and Applied Sciences 8 (3), 112-119 , 2014
    2014.0
    Citations: 3
  • content based hybrid DWT-DCT watermarking for image authentication in color images
    S Radharani, ML Valarmathi
    International Journal of Engineering Inventions 1 (4), 32-38 , 2012
    2012.0
    Citations: 2
  • Content based watermarking for color images using transform domain
    S Radharani, ML Valarmathi
    International Journal of Engineering Research and Applications 2 (1), 773-779 , 2012
    2012.0
    Citations: 1
  • Multiple watermarking scheme for image authentication and copyright protection using wavelet based texture properties and visual cryptography
    S Radharani, ML Valarmathi
    International Journal of Computer Applications 23 (3), 29-36 , 2011
    2011.0
    Citations: 26
  • A study on watermarking schemes for image authentication
    S Radharani, ML Valarmathi
    International Journal of Computer Applications 2 (4), 24-32 , 2010
    2010.0
    Citations: 58
  • Medical Image Watermarking
    S Radharani, M Umadevi, T Ramaprabha

MOST CITED SCHOLAR PUBLICATIONS

  • A study on watermarking schemes for image authentication
    S Radharani, ML Valarmathi
    International Journal of Computer Applications 2 (4), 24-32 , 2010
    2010.0
    Citations: 58
  • Multiple watermarking scheme for image authentication and copyright protection using wavelet based texture properties and visual cryptography
    S Radharani, ML Valarmathi
    International Journal of Computer Applications 23 (3), 29-36 , 2011
    2011.0
    Citations: 26
  • SCANNED DOCUMENT COMPRESSION USING HIGH EFFICIENCY VIDEO CODING (HEVC) STANDARD
    SR B.Nithya
    International Journal of Advances in Computer Science and Technology 3 (11 … , 2014
    2014.0
    Citations: 4
  • Content based Image Watermarking Scheme using Block SVD and Arnold Transform
    DMLV S. Radharani
    International Conference on Electronics and Communications Systems 1 (1 … , 2014
    2014.0
    Citations: 3
  • Content Based Watermarking Techniques using HSV and Fractal Dimension in Transform Domain
    S Radharani, ML Valarmathi
    Australian Journal of Basic and Applied Sciences 8 (3), 112-119 , 2014
    2014.0
    Citations: 3
  • content based hybrid DWT-DCT watermarking for image authentication in color images
    S Radharani, ML Valarmathi
    International Journal of Engineering Inventions 1 (4), 32-38 , 2012
    2012.0
    Citations: 2
  • Content based watermarking for color images using transform domain
    S Radharani, ML Valarmathi
    International Journal of Engineering Research and Applications 2 (1), 773-779 , 2012
    2012.0
    Citations: 1
  • Implementation of Digital Watermarking with Genetic Algorithm
    DS Radharani
    Sangam 13 (7-8), 153-159 , 2025
    2025.0
  • Study on Medical Image Watermarking
    DTR Dr. S. Radharani, M. Umadevi
    HTL JOURNAL 29 (9), 485 - 495 , 2023
    2023.0
  • Content-Based Watermarking Using MCA
    S Radharani, ML Valarmathi
    Proceedings of International Conference on ICT for Sustainable Development … , 2016
    2016.0
  • Content based Medical Image Watermarking Scheme Based on Arnold Transform with DWT Blending in Block SVD and Expansion for still images using ICA and Random blocks
    DMLV S. Radharani
    International Journal of Applied Engineering Research 10 (1), 1217-1227 , 2015
    2015.0
  • Medical Image Watermarking
    S Radharani, M Umadevi, T Ramaprabha