S GIRINATH

Scopus Publications

LARGE-SCALE GENOMIC PHENOTYPING OF RICE CROP HEALTH: INTEGRATING SENTINEL-2 NDVI COMPOSITES FOR CHARACTERIZED TRAIT SELECTION
S Girinath, V Anantha Natarajan
Genetics and Molecular Research, 2026
Crop management practices that helps to detect the presence of stress or any abnormality at the right time are essential to ensure food security and maximum agricultural output. The crop stress leads to massive loss of crop yield since it affects the plant health. Stress in crops is caused by poor irrigation, deficiency or any infection by pests. Modern agricultural methods manual scouting to determine the health condition of the crop is not efficient and not viable to large scale farming or large-scale monitoring of the area by the government agencies. This paper offers a dependable and effective method of determining and mapping stresses area within the farmed area. An unsupervised clustering framework that is based on deep learning was chosen to segment stressed area that provides valuable benefit as it can detect clusters of pixels that share certain common features. Unsupervised learning method does not presuppose previous knowledge or marked training data. It thus becomes an effective and scalable technique to study the health of crops, and it is also possible to map the levels of different stresses in the field without any problem. To begin with, an Auto-Encoder was applied to produce a small lower dimensional representation of the multi-band images. The AE model uses the abundant spectral information of several bands of Sentinel-2 to extract salient features that describe the underlying biophysical attributes of the crops. Then the encoded features are further grouped with k-Means clustering algorithm to outline the healthy vegetation, moderately stressed and stressed vegetation areas. The efficacy of the proposed approach was checked with the help of multiple performance measures that illustrate the effectiveness of this approach to offer a scalable solution to crop remote monitoring. This solution helps farmers to implement specific interventions and improve the management of the farms.
Hybrid Approach for Early Detection of Tomato Leaf Disease Using Advanced Machine Learning Algorithms
Modem Sai Kumar, Yadhav Balaji Rao, S. Girinath, Suresh Babu Jugunta
Smart Innovation Systems and Technologies, 2026
Unibridge-Revolutionizing Alumni Engagement
Sai Durga Penugonda, Kiran Kumar Jajala, Girinath S
Proceedings of 5th International Conference on Pervasive Computing and Social Networking Icpcsn 2025, 2025
Unibridge is an all-encompassing web-based alumni community platform aimed at building lifelong relationships between graduates and alma mater. The major goal of this project is to maximize alumni participation through an integrated system featuring key elements like alumni enrollment, job postings, mentorship, success stories monitoring, event attendance, and safe donation gateways. Built on python Django and SQL, the platform prioritizes ease of use, security, and scalability. System evaluation showed higher alumni engagement, enhanced institutional collaboration, and enhanced community involvement. Unibridge provides a realistic standard for contemporary digital alumni association platforms through career support promotion, knowledge sharing, and long-term institutional development.
Hemoglux Monitor: A Non-Invasive Glucose, Hemoglobin, and Pulse Monitor
Balambigai S, Sarankumar R, Kapilesh Kumar S, Girinath S, Sangeeth S, Nitheesh Kumar N
Proceedings of 8th International Conference on Trends in Electronics and Informatics Icoei 2025, 2025
Diabetes mellitus is a chronic metabolic disorder that requires regular monitoring of blood glucose levels to prevent severe health complications. Traditional glucose monitoring methods, such as finger pricking, cause discomfort and increase the risk of infections, which can hinder long-term management. This paper introduces the Hemoglux Monitor, a non-invasive device designed for real-time monitoring of glucose, hemoglobin, and pulse rate using Near-Infrared (NIR) spectroscopy and an ESP32 microcontroller. The device uses a 940 nm NIR LED and a photodiode to measure glucose levels through light absorption and reflection, transmitting the processed data wirelessly via Wi-Fi and Bluetooth to mobile devices for remote monitoring. Real-time glucose measurements are displayed on an internal LCD and continuously monitored via a mobile app. The system was evaluated on 10 participants, achieving an accuracy of 91.5%, with a mean absolute error (MAE) ranging from -7.2 mg/dL to 7.2 mg/dL compared to traditional invasive glucose meters. By eliminating the need for needle-based blood sampling, the Hemoglux Monitor provides a painless, cost-effective solution for diabetes management. The device integrates IoT and edge computing technologies to enhance data processing and enable continuous, remote health monitoring. The Hemoglux Monitor represents a significant step toward improving diabetes care through non-invasive technology, offering potential benefits for both clinical and home use.
Explainable AI-Powered Fraud Detection Device for Secure Financial Transactions
S. Mallikarjunaiah, E. Appa Rao, S. Girinath, Savithri H, V.S. Prasad Kandi, Maddikera Kalyan Chakravarthi
Proceedings Iceconf 2025 2025 2nd International Conference on Artificial Intelligence and Knowledge Discovery in Concurrent Engineering, 2025
The present paper describes an Explainable AIPowered Fraud Detection Unit to Secure Financial Transactions, which was created on the basis of Hybrid Deep Learning and Machine Learning Approach combined with Federated Learning. The proposed system is going to have a combination of Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks to identify complex patterns of fraud in transaction data, and Support Vector Machines (SVM) to be efficient when used with structured data. TensorFlow and TensorFlow Federated (TFF) will protect privacy and guarantee high performance since they allow decentralized training on client devices. Integration of Layer-wise Relevance Propagation (LRP) offers explainable and easily understandable results by improving the clarity of the model predictions. The system has an accuracy of 96, which is very high compared to the traditional techniques like random forest and SVM, thus it is very strong in detecting real-time fraud. Scalability issues and requirements on computational resources are also addressed in the paper, with some suggestions on how such problems can be improved in future deployments of large scale systems.
Enhancing Industrial Applications with LSTM-Based Predictive Analytics
Lalitha Ramachandran, S Girinath, N Padmavathi, G. Subramaniam, M. Siva Chitra, V Sathya
2024 15th International Conference on Computing Communication and Networking Technologies Icccnt 2024, 2024
Predictive analytics in industrial applications has the potential to significantly boost productivity and enhance operational efficiency. The research intends to solve is the difficulty of traditional models to capture the detailed temporal correlations and complex non-linear patterns that are common in industrial data. Long Short-Term Memory (LSTM)-based predictive analytics are the primary focus of this research, with the primary objective being the enhancement of industrial processes. Predicting complex and ever-changing patterns within operational data is a challenge in the modern industrial environment. The allocation of resources and decision-making processes are not as ideal as they may be. The LSTM architecture enables to use of the temporal correlations of the data in order to improve ability to predict the future. LSTM-based predictive analytics are significantly more accurate than conventional models when it comes to predicting the events that will occur in industrial processes. The findings shows LSTM-based predictive analytics have the potential to revolutionize industrial decision-making, which in turn leads to improved resource management and operational efficiency.
DeepCardioNet: Efficient Left Ventricular Epicardium and Endocardium Segmentation using Computer Vision
Bukka Shobharani, S Girinath, K. Suresh Babu, J. Chenni Kumaran, Yousef A.Baker El-Ebiary, S. Farhad
International Journal of Advanced Computer Science and Applications, 2024
In the realm of medical image analysis, accurate segmentation of cardiac structures is essential for accurate diagnosis and therapy planning. Using the efficient Attention Swin U-Net architecture, this study provides DEEPCARDIONET, a novel computer vision approach for effectively segmenting the left ventricular epicardium and endocardium. The paper presents DEEPCARDIONET, a cutting-edge computer vision method designed to efficiently separate the left ventricular epicardium and endocardium in medical pictures. The main innovation of DEEPCARDIONET is that it makes use of the Attention Swin U-Net architecture, a state-of-the-art framework that is well-known for its capacity to collect contextual information and complicated attributes. Specially designed for the segmentation task, the Attention Swin U-Net guarantees superior performance in identifying the relevant left ventricular characteristics. The model's ability to identify positive instances with high precision and a low false positive rate is demonstrated by its good sensitivity, specificity, and accuracy. The Dice Similarity Coefficient (DSC) illustrates the improved performance of the proposed method in addition to accuracy, showing how effectively it captures spatial overlaps between predicted and ground truth segmentations. The model's generalizability and performance in a variety of medical imaging contexts are demonstrated by its application and evaluation across many datasets. DEEPCARDIONET is an intriguing method for enhancing cardiac picture segmentation, with potential applications in clinical diagnosis and treatment planning. The proposed method achieves an amazing accuracy of 99.21% by using a deep neural network architecture, which significantly beats existing models like TransUNet, MedT, and FAT-Net. The implementation, which uses Python, demonstrates how versatile and useful the language is for the scientific computing community.
Fuzzy Integrated Latent Dirichlet Allocation Algorithm for Intrusion Detection in Cloud Environments
International Journal of Intelligent Systems and Applications in Engineering, 2024
Real-Time Identification of Medicinal Plants Using Deep Learning Techniques
S. Girinath, Peddinti Neeraja, M. Sunil Kumar, S. Kalyani, B. Lakshmi Mamatha, N.R.T. GruhaLakshmi
Tqcebt 2024 2nd IEEE International Conference on Trends in Quantum Computing and Emerging Business Technologies 2024, 2024
A significant portion of the global population uses medicinal plants as a primary source of therapeutic chemicals and as an alternate source of pharmaceuticals. Plant identification from photos is a rapidly developing field of research at the moment due to recent developments in computer vision. High precision, accuracy, and practical application were demonstrated by several findings. Exact automatic identification of medicinal plants was the aim of this study. The illumination in the area is uneven, making it challenging to distinguish a leaf from its backdrop. We provide a procedure that uses a sample of a plant's leaves to identify the species. We provide a new collection of medicinal plant photos, which includes one and ten (10) different plant species categories. In the suggested technique, relevant information is automatically extracted from plant photos using convolutional neural networks (CNNs). Large-scale high-resolution photographs of a variety of therapeutic plant species are used to train the CNN model. Using distinguishing visual characteristics, such as leaf form, color, and texture, the trained model is able to identify several plant species. Utilizing a smartphone or other portable device, users may snap images of plants with the developed system and share them with others for real-time plant identification on a mobile or embedded platform. Following analysis of the obtained image using the trained CNN model, a precise classification result is obtained fast.
Exploring K-Means Meta-Heuristic Techniques For Prediction of Anomalies In IoT-Enabled Industrial Systems
T. Gnanasekaran, S Girinath, K Venkatesh, N. Valarmathi, Sunil Kumar Bandili, S Balasubramani
2024 15th International Conference on Computing Communication and Networking Technologies Icccnt 2024, 2024
Due to the broad implementation of the Internet of Things (IoT) in industrial systems, the introduction of Industry 4.0 resulted in a significant increase in operational efficiency. However, it also resulted in the introduction of new challenges when it came to identifying abnormalities. This paper presents a new method for boosting the flexibility and precision of anomaly detection models by integrating the best aspects of K-Means clustering with meta-heuristic approaches. The utilisation of K-Means meta-heuristic techniques is the driving force behind this research, which is being driven by the growing demand for reliable anomaly prediction in industrial systems that are enabled by the IoT. Aiming to discover methods that can improve the reliability of anomaly detection algorithms is the objective of this research. K-Means clustering is systematically integrated with metaheuristic techniques in this study, which allows for the optimisation of the clustering algorithm for the purpose of detecting abnormalities in industrial data that is enabled by the IoT. We use real-world datasets from a wide range of businesses in order to measure the effectiveness of the proposed strategy in comparison to more traditional ways. The results demonstrate that the K-Means meta-heuristic strategy greatly enhances the accuracy of anomaly detection in comparison to the approaches that are considered conventional. Recall, false positive rates, and precision are all areas in which it excels beyond its capabilities. The results indicate that this unique approach has the potential to be useful when applied to the challenges of anomaly detection in industrial systems that are enabled by the IoT.
Crop Yield Prediction Using Machine Learning
M. Sunil Kumar, S. Girinath, G. Guru Venkata Siva Lakshmi, A. Venkata Shiva Ganesh, K. Jayanth Kumar
2023 International Conference on Sustainable Emerging Innovations in Engineering and Technology Icseiet 2023, 2023
Plant Disease Identification Tracking and Forecasting Using Machine Learning
M. Sowmya Vani, S. Girinath, V. Hemasree, Lomada Hars Havardhan, Poli Sandhya
Proceedings International Conference on Technological Advancements in Computational Sciences Ictacs 2023, 2023
Enhancing Sentiment Analysis with an AttentionBased Machine Learning Model
M. Sunil Kumar, S. Girinath, A. Mohitha, K. Ramyasree, K. Mounika
Proceedings International Conference on Technological Advancements in Computational Sciences Ictacs 2023, 2023
Deep Learning-based Segmentation and Computer Vision-based Ultrasound Imagery Techniques
S Girinath, T Kavitha, Pamulapati Satish Chandra, Nellore Manoj Kumar, Naveen Kishore Gattim, Ratna Kumari Challa
2023 5th International Conference on Electrical Computer and Communication Technologies Icecct 2023, 2023

S GIRINATH

RESEARCH, TEACHING, or OTHER INTERESTS

Scopus Publications

RECENT SCHOLAR PUBLICATIONS

MOST CITED SCHOLAR PUBLICATIONS