VETRISELVI T

Dr. VETRISELVI T


Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
29 January 2023 in Materials
Verified
Journal Article
Open access
PDF file
Pushpa GothwalPushpa Gothwal
Pushpa Gothwal
AKAjay Kumar
Ajay Kumar
Deepshikha RathoreDeepshikha Rathore
Deepshikha Rathore
RMRana Mukherji
Rana Mukherji
VTVETRISELVI T
VETRISELVI T
ASAnandan Sambandam
Anandan Sambandam
This paper presents an experimental analysis of the optimization of PZT-based tiles for energy harvesting. The hardware (actual experiment), PZT-based tiles, were developed using 6 × 6 piezoelectric (PZT—lead zirconate titanate) sensors of 40 mm in diameter on a hard cardboard sheet (300 × 300 mm2). Our experimental analysis of the designed tiles obtained an optimized power of 3.626 mW (85 kg or 0.83 kN using 36 sensors) for one footstep and 0.9 mW for 30 footsteps at high tapping frequency. Theoretical analysis was conducted with software (Design-Expert) using the response surface methodology (RSM) for optimized PZT tiles, obtaining a power of 6784.155 mW at 150 kg or 1.47 kN weight using 34 sensors. This software helped to formulate the mathematical equation for the most suitable PZT tile model for power optimization. It used the quadratic model to provide adjusted and predicted R2 values of 0.9916 and 0.9650, respectively. The values were less than 0.2 apart, which indicates a high correlation between the actual and predicted values. The outcome of the various experiments can help with the selection of input factors for optimized power during pavement design.

1 January 2025 in International Journal of Neutrosophic Science
Verified
Journal Article
Closed access
PDF file
MSM. Sivakumar
M. Sivakumar
ABAbdul Rajak Rabıyathul Basarıya
Abdul Rajak Rabıyathul Basarıya
MSM. Senthil
M. Senthil
VTVETRISELVI T
VETRISELVI T
GRG. Raja
G. Raja
RRR. Rajavarman
R. Rajavarman
American scientific publishing group

12 July 2023 in E3S Web of Conferences
Verified
Conference Paper
Open access
PDF file
VTVETRISELVI T
VETRISELVI T
MMMihir Mathur
Mihir Mathur
Automatic text summarization is increasingly required with the exponential growth of unstructured text through increasing internet and social media usage across the globe. The various approaches are outcomes of extraction-based and abstraction-based. In Extraction-based summarization, the extracted content from the original data, is typically presented in the same or slightly modified form without significant paraphrasing or restructuring. Abstractive methods involve building an internal representation of the original content and then using that representation to generate a summary that may not be present in the original text verbatim. They employ natural language processing techniques, such as language generation models, to paraphrase and rephrase sections of the source document to create a more human-like summary. Abstraction in abstractive summarization is indeed a challenging task because it requires not only linguistic and syntactic understanding but also a deeper semantic understanding of the original content. The evaluation of automatically summarized text through criteria of precision and recall, and various evaluation methods, datasets used for domain based and generic text summarization. This paper proposes the Sentence Length Impact (SLI) algorithm to summarize English text content, which gives 92% accuracy and translating the same in French.

11 June 2026 in Scientific Reports
Verified
Journal Article
Open access
PGPriyadharshini G
Priyadharshini G
VTVETRISELVI T
VETRISELVI T
During the early growth stages of crops, the weeds present can severely affect crop yields. Accurate and fast identification of the types of weeds present using robots can help mitigate this problem, allowing agronomists and farmers to employ more precise methods to control weed growth. In this research, we designed the custom Lightweight CNN-SE-Transformer architecture with smaller computational cost while still being able to capture a wide range of important features. This research evaluates the various convolutional network depth by increasing and decreasing the CNN layers. The model was evaluated with residuals and without residuals, skip connections to learn the ability of light weight architecture. Squeeze-and-Excitation (SE) modules in the convolutional backbone to increase the networks discriminative ability. The sequence of multiple convolutional layers with residuals helps to extract the deep feature extraction which is necessary dealing with small plants with more similarities. The main difference from previous deep learning methods is that the model is able to shrink the sequence of tokens and utilize the subsequent features to more efficiently analyze the data while still maintaining the overall performance of the model. To explicitly prove the proposed model ability, model was evaluated with two strong datasets with three class and 16 class weeds. Proposed architecture performed well on both datasets and achieves best test accuracy. For the three optimizers, Adam, RMSProp, and adam+label smoothing, the performance of each optimizer is evaluated to gain insight on the convergence and stability of the model under a wide variety of training conditions. The results across the three models have shown that Adam optimizer achieves 97% of test accuracy was the best optimizer while Adam label smoothing achieves similar to 96.5% and RMSProp achieves 95% of test accuracy shows less performance. Explainability of model in gradcam shows adam with proposed custom lightweight model understand the exact features of weed.