Vinayak Gupta

Mr. Vinayak Gupta

Computer Science and Engineering, ITS Engineering College, Greater Noida, India
+1
orcid

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
24 February 2025 in Journal of Imaging
Verified
Journal Article
Open access
PDF file
YDYajnaseni Dash
Yajnaseni Dash
Vinayak GuptaVinayak Gupta
Vinayak Gupta
Ajith AbrahamAjith Abraham
Ajith Abraham
SCSwati Hochschule Heidelberg Chandna
Swati Hochschule Heidelberg Chandna
The advancement of technology has ushered in remote sensing with the adoption of high-altitude infrared thermal object detection to leverage the distinct advantages of high-altitude platforms. These new technologies readily capture the thermal signatures of objects from an elevated point, generally unmanned aerial vehicles or drones, and thus allow for the enhancement of the detection and monitoring of extensive areas. This study explores the application of YOLOv8’s advanced architecture, as well as dynamic magnitude-based pruning techniques paired with non-maximum suppression for high-altitude infrared thermal object detection using UAVs. The current research addresses the complexities of processing high-resolution thermal imagery, where traditional methods fall short. We converted dataset annotations from the COCO and PASCAL VOC formats to YOLO’s required format, enabling efficient model training and inference. The results demonstrate the proposed architecture’s superior speed and accuracy, effectively handling thermal signatures and object detection. Precision–recall metrics indicate robust performance, though some misclassification, particularly for persons, suggests areas for further refinement. This work highlights the advanced architecture of YOLOv8’s potential in enhancing UAV-based thermal imaging applications, paving the way for more effective real-time object detection solutions.

24 June 2024
Verified
Conference Paper
Closed access
Vinayak GuptaVinayak Gupta
Vinayak Gupta
RSRicha Sharma
Richa Sharma
Breast cancer poses a significant health threat to the growing population of the world and as such calls for innovative methods to improve early detection and improve patient safety. The existing techniques in the field have made use of techniques like various image processing models like RetinaNet, You Only Look Once, Convolutional Neural Networks as well as tree-based models like the J48, decision trees and Naïve Bayes to predict Brest cancer but most these studies focus on the accuracy of the predictions without providing a basis for explanation and understandability of the results. Therefore, to provide transparency, this study utilizes the Local Interpretable Model-agnostic Explanations (LIME). This study investigated the performance of three boosting algorithms LightGBM, CatBoost and XGBoost in breast cancer prediction making use of the SEER dataset. The paper also makes use of SMOTE and the amalgamation of Tomek Links to handle class imbalance generally present with medical data to enhance the generalizability of the results. With hyperparameter tuning, the results show that LightGBM outperformed the other two in terms of accuracy as well as recall metrics. LIME explanations were then employed specific patient instances to understand the interplay of the various variables and how they contribute to the model predictions. These explanations highlighted the clinical relevance of variables such as T Stage and Grade of cancer, aiming to instill confidence among the clinicians. The research in all undermines the importance of integrating XAI methodologies in breast cancer detection, paving way for more clinically regulated and relevant models in the future.

14 March 2024
Verified
Conference Paper
Closed access
Vinayak GuptaVinayak Gupta
Vinayak Gupta
STShailja Tripathi
Shailja Tripathi
DSDhruv Singh
Dhruv Singh
ABAbhay Bansal
Abhay Bansal
Postpartum depression (PPD) is a growing concern for mothers on a global stage and is usually linked to the varied emotional changes which happen to a woman postdelivery. This issue is a pressing one and as such early detection acts as the essential bridge between growth and development of the mother-child bond while promoting a nurturing environment. The paper makes use of two classification algorithms namely, CatBoost and LightGBM, on a dataset of 1503 records, with the primary aim to list out the various indicators which contribute to PPD. It has been seen that varied emotional indicators such as guilt, anger, sleep depravity and irritability act as the prime indicators of this disease. While comparing the two algorithms CatBoost outshines LightGBM owing to its prowess in handling categorical data and its ordered boosting approaches. In all this study outlines the potential of these algorithms in predictive modelling as well as timely detection of this disease, establishing the foundation for better efforts to mitigate and manage PPD.

4 April 2025
Verified
Conference Paper
Closed access
Vinayak GuptaVinayak Gupta
Vinayak Gupta
NMNaman Mishra
Naman Mishra
YDYajnaseni Dash
Yajnaseni Dash
UKUpendra Kumar
Upendra Kumar
AAAjith Abraham
Ajith Abraham
With a growing need to tackle the issue of fraudulent transactions in the financial domain, especially with the boom of cryptocurrency. This study proposes a Graph Convolutional Network (GCN)--based classification model for detecting fraudulent transactions by leveraging graph-structured data. Traditional classification methods often fail to capture complex topological dependencies within financial networks, whereas GCNs effectively learn from node connectivity and feature propagation. The dataset undergoes extensive preprocessing, including KNN imputation, power transformations, polynomial feature engineering, and variance thresholding, ensuring optimal feature selection. To address the class imbalance, a hybrid SMOTE-ADASYN resampling technique is employed. Experimental results demonstrate that the proposed GCN outperforms CatBoost, XGBoost, SVM, Logistic Regression, and Random Forest, achieving an accuracy of 98.75% and an AUC score of 0.97. Additionally, subgraph visualizations provide interpretability into model predictions. This work establishes GCNs as a superior alternative for fraud detection in transaction networks.

25 April 2025
Verified
Book Chapter
Closed access
PDF file
Vinayak GuptaVinayak Gupta
Vinayak Gupta
YDYajnaseni Dash
Yajnaseni Dash
SSSudhir C. Sarangi
Sudhir C. Sarangi
Ajith AbrahamAjith Abraham
Ajith Abraham

16 May 2025 in IEEE Access
Verified
Journal Article
Open access
YDYajnaseni Dash
Yajnaseni Dash
SSSudhir C. Sarangi
Sudhir C. Sarangi
Vinayak GuptaVinayak Gupta
Vinayak Gupta
NKNaween Kumar
Naween Kumar
Ajith AbrahamAjith Abraham
Ajith Abraham
The present work describes a new method for detecting lung cancer based on a Kolmogorov-Arnold Network (KAN) combined with a Context-Aware Feature Pyramid Network (FPN), denoted KAFPN and supported by XAI tools, including Local Interpretable Model-agnostic Explanations (LIME) and SHapley Additive exPlanations (SHAP) for explanation purposes. The model is compiled based on an extended corpus containing 25,000 histopathological images from 15,000 lung tissue samples. The proposed KAFPN model outperformed the traditional models, such as Base CNN, Inception V3, VGG16, ResNet50, EffecientNet-B7, ResNet-101, and DenseNet-121, based on testing and validation accuracy of 99.63% and 99.60%, respectively. The integration of LIME and SHAP facilitated an advanced level of interpretability by providing detailed visual explanations of the model’s predictive decisions, delineating the image regions that most significantly impacted the decision-making process. The proposed KAFPN framework is well suited for image segmentation and object detection because it has multiple-scale feature responses. Stringent data augmentation and correct division of data sets make the model robust and efficient for real-world applications.