Computer Science, Multidisciplinary, Agronomy and Crop Science, Computer Vision and Pattern Recognition
17
Scopus Publications
120
Scholar Citations
5
Scholar h-index
2
Scholar i10-index
Scopus Publications
Hybrid Morphological U-Net with a Spectral Morphological Transformer for Superior Plant Leaf Diseases Classification S. Gopinath, M. Madheswaran, S. Arularasi, R. Mohanabharathi International Journal of Advanced Science and Engineering, 2026 In agriculture, plant diseases severely affect productivity and lead to starvation.Hence, early treatment of plant diseases is crucial.However, the existing disease detection approaches are time consuming and require human intervention.To overcome these issues, deep learning models have been extensively used.The primary objective of this study is to develop an automated and efficient hybrid deep learning model by combining a Morphological U-Net, a Convolutional Neural Network (CNN) and a Spectral Morphological Transformer, a Vision Transformer (ViT) to perform accurate plant diseases detection.This hybrid model combines local feature extraction and global contextual learning to improve lesion detection and localization and classification accuracy.To analyse the effectiveness of the hybrid model, the benchmark PlantVillage dataset is used.Among the multiple plant species in this dataset, pepper, tomato, and potato plant leaves and their corresponding classes are selected for training and testing.The proposed model accurately classified multiple class plant diseases with an accuracy of 97.89% and outperformed existing deep learning models.The significance of this study lies in timely intervention for plant diseases, reducing crop loss, and promoting precision agriculture practices.It is apparent that this model is suitable for challenging environmental condition and edge computing deployment.
IoT-Enabled Wearable Healthcare Device with Real-Time ECG Monitoring and Cloud Analytics Saravanan V, Selvamani Indrajith, Nisha J C, Kalaiyarasi D, Gopinath S, Srilakshmi K Ssrg International Journal of Electronics and Communication Engineering, 2025 The conventional ECG methods used to monitor vital signs are limited by their reliance on hospital equipment, restricted accessibility, and the late onset of diagnosis. To overcome these obstacles, an IoT-based wearable healthcare device is suggested to track the real-time ECG and analytics in the cloud. The product combines wearable ECG, SpO2, and heart rate variability sensors with an IoT microcontroller, backed by optimized communication protocols and cloud storage. A hybrid CNN-LM Deep Learning Model, based on arrhythmia classification, is employed, and mathematical models are utilized to compare energy efficiency and latency. In experimental testing, an accuracy of 98.6%, a sensitivity of 97.9%, a specificity of 98.2%, a precision of 98.3%, a F1-score of 98.1%, an average latency of 45 ms, a packet delivery ratio of 99.2%, and an energy consumption of 18.7 mW were achieved. These findings support the efficiency of the developed system in providing scalable, energy-efficient, and accurate real-time cardiac monitoring to support innovative healthcare applications.
Enhanced Livestock Monitoring with Modified Kalman Filter and Decision Tree Algorithm for Noise-Reduction in Sensor Data Arun C A, Sudhan M B, Thripthi P Balakrishnan, Gopinath S, Srividhya N, Saravanan V, Navaneethan S Ssrg International Journal of Electrical and Electronics Engineering, 2025 Kalman filtering, a robust statistical estimation method, has emerged as a pivotal tool across disciplines, excelling in noise reduction and state estimation with minimal computational demands. Its adaptability has fostered diverse implementations, notably in complex sensory and robotic systems. This technique significantly enhances system reliability and efficiency by effectively filtering signals from noise, thereby catalyzing technological progress across numerous sectors. The method is precious in modern systems that rely on high-sensitivity sensors like accelerometers and gyroscopes, which are crucial for improved performance but vulnerable to noise. By addressing the challenge of noisy data readings, which can significantly impact system performance. By strategically deploying sensors on cattle, real-time data on animal movements are collected, and with this, we can predict valuable insights into their daily activities are possible. This innovative research on animal welfare management has been adopted to optimize farm operations. The suggested revised algorithm of the standard Kalman filter can effectively minimize noise in the livestock management system. Different combinations of values for the Q and R variables are tested and tabulated at Q = 0.01 and R = 100, also Q = 0.01, and R = 1000 we got maximum results. Also, we get better results on the proposed modified Kalman filter with the decision tree algorithm, which effectively predicts the actual data with an accuracy of 88.67%, precision of 87.53%, recall of 87.5%, and F1 score of 87.47%, indicating its strong ability to capture the underlying patterns in the dataset. In contrast, the Linear Regression model may be underfitting due to its inability to model such non-linearity effectively. When comparing the results, the decision tree regression method outperformed linear regression and polynomial regression methods. Also, it is well-suited for capturing sudden shifts and plateaus in the cattle's behavior. This innovative project adopted an advanced system that enhances animal welfare management and optimizes farm operations.
Personalized Recommendations System for Mental Health Support using NLP Akash G, Gopinath S, Mohamed Zachariah S M, S. Mithuna 3rd International Conference on Electronics and Renewable Systems Icears 2025 Proceedings, 2025 The prevalence of mental health conditions is increasing so much that detecting them early will make interventions more effective. I introduce a recommendation system which benefits from Natural Language Processing combined with machine learning approaches to analyze users’ mental health conditions through their written and spoken content. Users can transmit writing or speech data directly to the system for processing. Among all available algorithms Machine learning algorithms including Logistic Regression, Random Forest, Decision Tree and Multinomial Naive Bayes exhibit the best performance. The system performs user classification based on their diagnosed mental disorder which enables it to present appropriate mental health video content specifically designed to support their needs. Their goal was to create a system which provides real-time feedback while raising accessibility to mental health care and starting interventions early. Integration of advanced technology into personal mental health experiences enters a critical phase because of these developments.
Real-Time Multi-Modal Deepfake Detection Using Spatiotemporal Attention Networks and Cross-Domain Adversarial Training A P. Gopu, R. Rakesh, P. Swetha, P. Tharun, S. Gopinath, K. Baby 2025 1st International Conference on Advancement in Futuristic Technologies Icaft 2025, 2025 The CNN-LSTM structures implemented for deepfake detection approaches have displayed effectiveness, but their accuracy has been limited and cross domain success has been limited due to less than optimal spatiotemporal modeling and ability to detectdeepfakes in real time. To improve upon these limitations, our work proposes MS-STAN (Multi-Scale Spatiotemporal Attention Network), a spatiotemporal framework for deepfake detection that utilizes attention mechanisms to improve understanding of spatial features and temporal coherence. To accomplish this, the proposed model integrates EfficientNetV2 for enhanced extraction of spatial features, Bi-directional LSTMs to better exploit temporal dependencies, and self-attention to better aggregate temporal dependencies. To reinforce generalization by training across domains and without further delay, cross-domain adversarial training is also employed. Experimental findings on FaceForensics++, Celeb-DF, DFDCshow a significant improvement in accuracy and real-time detection at 45 FPS with an accuracy of 77%.In addition, our proposed spatiotemporal detection implements heatmaps to visualize spatiotemporal features that contributed to the decision in the classification process.
Deep Learning-Based Approach for Accurate Plant Disease Identification Using Image Analysis K. Sakthivel, S. Arularasi, S. Gopinath, M. Vinoth, G. Kowsalya, S. Lalitha Proceedings 3rd International Conference on Artificial Intelligence and Machine Learning Applications Healthcare and Internet of Things Aimla 2025, 2025 Plant ailments pose an important warning to allencompassing food safety, provoking solid financial deficits and negotiating the livelihoods of millions of population general. Timely and correct labeling of plant ailments is important for achieving effective affliction administration planning. This study intends a deep knowledge-located approach for plant disease labeling utilizing concept study. A convolutional interconnected system (CNN) model was prepared on a dataset of countenances of vigorous and unnatural plant leaves. The model reached a veracity of 95.6% in labeling five coarse plant afflictions. The results demonstrate the influence of the projected approach inaccurate plant ailment labeling. The study donates to the incident of automated plant affliction disease arrangements, that maybe secondhand by farmers, land specialists, and added collaborators to better crop yields and decrease business-related losses.
Eye Detection Wheelchair with Smart Monitoring Using Deeplearning in CNN R. Muthuchelvan, Divagarvasu. P, Gopinath. S, Gokul Prasanth. V, Gowtham. J 2nd IEEE International Conference on Iot Communication and Automation Technology Icicat 2024, 2024 Individuals with severe cervical spinal cord injuries often face important challenges in interacting with their environment, including difficulties using smart phones or operating powered wheel chairs. For those with tetraplegia, Eye-tracking technologies have emerged as a crucial tool for enhancing control and convey. However, confinement in accuracy, practicality, an calibration time often plague conventional eye-tracking devices. Researchers have explored Convolutional Neural Networks(CNNs) as a potential solution to address these shortcoming in AI-powered eye tracking system. CNNs a type of deep learning architecture excel at recognizing complex pattern in a data, leading to more precise and reliable eye tracking. An mostly good way to argument the eye tracking correctness method. This technique outline a more reliable and result way to interact with digital mechanism by using deep learning algorithms to admit and track eye movements based on blink count. This technology has the power to conversion the way we interact with digital technologies, greatly improving their usefulness and accessibility for those who suffer from confusion or impairments. For human beings who have trouble with standard eye tracking because of motor restrictions, vision depletion, or other reductions
Ethnicity Prediction From Audio Samples for Different Languages Using Convolutional Neural Sharvaani Ravikumar Thoguluva, Suraj Gopinath, Udith S, Rimjhim Padam Singh, Sneha Kanchan 2024 5th International Conference for Emerging Technology Incet 2024, 2024 “Ethnicity Prediction from Voice Samples” deals with prediction of a person's ethnicity from his/her voice samples. Ethnicity Prediction has gained more and more importance and attention in fields like security, recommendations systems, learning apps etc. The proposed work uses a customized Convolutional Neural Network (CNN) based model to predict the ethnicity of the speaker based on combination of varied audio features like Mel-Frequency Cepstral Coefficients (MFCC), pitch, energy, and zero crossing rate. The different voice samples are obtained from the popular Speech Accent Archive Dataset. Augmentation of data by performing time and pitch shifting proved to give a significant boost in the accuracy as it increased the number of voice recordings. The work also compares the performance of the proposed model with other state-of-art models like LSTM. The F-score obtained was 98% for the class Russian, Polish and 90% for the class French, Italian, Spanish i.e. Romance Languages and 80% for mixed language category that comprises of Indian, Italian, German and Polish.
Emotion Recognition Using Meta-Learning based on Facial Expressions Prashanth V. J., Sri Naga Jathin P., Suraj Gopinath, Udith S., Kavitha C. R 2024 5th IEEE Global Conference for Advancement in Technology Gcat 2024, 2024 Emotion recognition is essential in many real-life problems and applications such as computer-human interaction, health monitoring, etc. In this project, we propose a meta-learning approach for developing a computer visionbased emotion recognition system. The aim of the work is to find human emotions from visual inputs and sort them accordingly. Meta learning techniques are used to allow the model to quickly adapt itself to emotion patterns using little training data that is short and compact. Therefore, meta learning technique is important since it allows the algorithm to perform well among different datasets with diverse data sets including unseen emotions (unknown emotions). Emotion recognition should address generalization problems while having ability to handle adaptation issues. It therefore seeks transparency in emotion recognition in addition to overcoming challenges of adaptability. Its objective is to offer answers to practical cases where emotion detection can be applied within dynamic environments. Thus at the end of this project we expect a reliable, faster, more dynamic way of classifying and recognizing human emotions based on visual inputs which highly valuable in understanding human emotional state for multiple other applications.
Resume Analyzer and Skill Enhancement Recommender System Prashanth V J, Sri Naga Jathin P, Suraj Gopinath, Udith S, Kavitha C.R. 2024 Asia Pacific Conference on Innovation in Technology Apcit 2024, 2024 The research "Resume Parsing and Job Prediction" mainly deals with prediction of an appropriate job title for a person based on their resume. Due to rapid increase in job applicants, companies find it difficult to manually read every resume as this work requires tremendous amounts of time and effort. Hence the resume parser has gained a lot of importance in recent years. Here we use a custom CNN model to predict the job role suitable for an applicant based on their resume. We also compare the model’s performance with other machine learning models Random Forest and SVM and neural network models like custom CNN and a pre trained BERT model. Job applicants also find it difficult to check how compatible they would be for a job. Hence, we use a word2vec vectorization technique and cosine similarity and find the similarity between the resume and a company’s job description. This application also includes a scoring and a feedback system that provides personalized feedback for the applicant’s resume and scores it based on important contents like their education, skills, experience.
Image Processing Based Automatic Traffic Control System Sudhakiran Ponnuru, Ambrish Kumar Sharma, Patan Feridoz Khan, E Poovannan, B Venkata Lakshmi, S Gopinath Proceedings of the 2024 International Conference on Innovative Computing Intelligent Communication and Smart Electrical Systems Icses 2024, 2024
Survey and challenges of Li-Fi with comparison of Wi-Fi P. Kuppusamy, S. Muthuraj, S. Gopinath Proceedings of the 2016 IEEE International Conference on Wireless Communications Signal Processing and Networking Wispnet 2016, 2016
RECENT SCHOLAR PUBLICATIONS
Machine Learning Model Based Urban Temperature Analysis with Fuzzy Reinforcement Neural Network. L Pallavi, G Shravani, JS Devi, BS Lakshmi, M Pushpalatha, S Gopinath, ... Journal of Intelligent Systems & Internet of Things 18 (1) , 2026 2026
Cloud IoT with Remote Sensing Data Segmentation and Classification Using Deep Learning Model for Sustainable Agriculture. T Shanmugapriya, RM Rani, GR Babu, T Srinivasulu, S Saranya, ... Journal of Intelligent Systems & Internet of Things 18 (1) , 2026 2026 Citations: 1
Sustainable crop recommendation system using seasonally adaptive recursive spectral convolutional neural network for responsible agricultural production G Selvaraj, S Kuppusamy, M Aswathanarayanan Geomatics, Natural Hazards and Risk 16 (1), 2509619 , 2025 2025 Citations: 18
LCNFN: LeNet‐Cascade Neuro‐Fuzzy Network for Grape Leaf Disease Segmentation and Multi‐Classification G Selvaraj, SV Puthenkaleelkal, P Alaguchamy, STN Senthil Journal of Phytopathology 173 (3), e70061 , 2025 2025 Citations: 1
Deep Learning-Based Approach for Accurate Plant Disease Identification Using Image Analysis K Sakthivel, S Arularasi, S Gopinath, M Vinoth, G Kowsalya, S Lalitha 2025 3rd International Conference on Artificial Intelligence and Machine … , 2025 2025 Citations: 8
Image Processing Based Automatic Traffic Control System S Ponnuru, AK Sharma, PF Khan, E Poovannan, BV Lakshmi, S Gopinath 2024 International Conference on Innovative Computing, Intelligent … , 2024 2024
Auguring and Predictions of COVID-19 by ML and PowerBI BG Geetha, K Sakthivel, S Gopinath, G Kowsalya 2023 9th International Conference on Smart Structures and Systems (ICSSS), 1-4 , 2023 2023
A detection of amblyopia using image processing and machine learning techniques K Praveen, S Lalitha, S Gopinath J. Comput. Sci. Eng. Softw. Test. 5 (2), 1-11 , 2023 2023 Citations: 1
A novel approach for IoT intrusion detection system using modified optimizer and convolutional neural network S Vijayalakshmi, TD Subha, ES Reddy, D Yaswanth, S Gopinath 2022 Sixth International Conference on I-SMAC (IoT in Social, Mobile … , 2022 2022 Citations: 6
RETRACTED: A plant disease image using convolutional recurrent neural network procedure intended for big data plant classification S Gopinath, K Sakthivel, S Lalitha Journal of Intelligent & Fuzzy Systems 43 (4), 4173-4186 , 2022 2022 Citations: 9
A detection of amblyopia medical condition in biomedical datasets using image segmentation and detection processing S Lalitha, N Shanthi, S Gopinath Journal of Medical Imaging and Health Informatics 11 (11), 2814-2821 , 2021 2021 Citations: 1
Automated Vision Defect Detection Supported Deep Convolutional Neural Networks S Lalitha, N Shanthi, S Gopinath Journal of Physics: Conference Series 1964 (4), 042044 , 2021 2021
Design and implementation of a deep convolutional neural networks hardware accelerator K Sekar, S Gopinath, K Sakthivel, S Lalitha Journal of Physics: Conference Series 1964 (5), 052008 , 2021 2021 Citations: 3
PLANT DISESASE DETECTION BY USING IMAGE PROCESSING TECHNIQUES LS GOPINATH S IJRIET 3 (08), 1-5 , 2017 2017 Citations: 3
LOGICAL MODEL TO DISCOVER SPITEFUL NODES IN MANETS gopinath s International Journal of Current Trends in Engineering & Research 2 (5), 676-684 , 2016 2016
Survey and challenges of Li-Fi with comparison of Wi-Fi P Kuppusamy, S Muthuraj, S Gopinath 2016 International Conference on Wireless Communications, Signal Processing … , 2016 2016 Citations: 68
A PRIVACY-PRESERVING ACCESS CONTROL WITH ROBUST DATA AUTHENTICITY FOR CLOUD GROUP P Sathya, S Gopinath, P Kuppusamy IJETCSE 19 (1), 35-40 , 2015 2015 Citations: 1
Adaption for Improved Lifetime in Wireless Sensor Networks with Graph Factorization G S International Journal Of Infinite Innovations In Engineering And Technology … , 2015 2015
Privacy Preserving and Truthful Detection of Packet Dropping Attacks in Wireless Adhoc Networks G S International Journal of Science and Engineering Research 3 (10) , 2015 2015
Geographic Routing Based Adaptive Location Update for Mobile Adhoc Networks International Journal Of Infinite Innovations In Engineering And Technology … , 2015 2015
MOST CITED SCHOLAR PUBLICATIONS
Survey and challenges of Li-Fi with comparison of Wi-Fi P Kuppusamy, S Muthuraj, S Gopinath 2016 International Conference on Wireless Communications, Signal Processing … , 2016 2016 Citations: 68
Sustainable crop recommendation system using seasonally adaptive recursive spectral convolutional neural network for responsible agricultural production G Selvaraj, S Kuppusamy, M Aswathanarayanan Geomatics, Natural Hazards and Risk 16 (1), 2509619 , 2025 2025 Citations: 18
RETRACTED: A plant disease image using convolutional recurrent neural network procedure intended for big data plant classification S Gopinath, K Sakthivel, S Lalitha Journal of Intelligent & Fuzzy Systems 43 (4), 4173-4186 , 2022 2022 Citations: 9
Deep Learning-Based Approach for Accurate Plant Disease Identification Using Image Analysis K Sakthivel, S Arularasi, S Gopinath, M Vinoth, G Kowsalya, S Lalitha 2025 3rd International Conference on Artificial Intelligence and Machine … , 2025 2025 Citations: 8
A novel approach for IoT intrusion detection system using modified optimizer and convolutional neural network S Vijayalakshmi, TD Subha, ES Reddy, D Yaswanth, S Gopinath 2022 Sixth International Conference on I-SMAC (IoT in Social, Mobile … , 2022 2022 Citations: 6
Design and implementation of a deep convolutional neural networks hardware accelerator K Sekar, S Gopinath, K Sakthivel, S Lalitha Journal of Physics: Conference Series 1964 (5), 052008 , 2021 2021 Citations: 3
PLANT DISESASE DETECTION BY USING IMAGE PROCESSING TECHNIQUES LS GOPINATH S IJRIET 3 (08), 1-5 , 2017 2017 Citations: 3
Cloud IoT with Remote Sensing Data Segmentation and Classification Using Deep Learning Model for Sustainable Agriculture. T Shanmugapriya, RM Rani, GR Babu, T Srinivasulu, S Saranya, ... Journal of Intelligent Systems & Internet of Things 18 (1) , 2026 2026 Citations: 1
LCNFN: LeNet‐Cascade Neuro‐Fuzzy Network for Grape Leaf Disease Segmentation and Multi‐Classification G Selvaraj, SV Puthenkaleelkal, P Alaguchamy, STN Senthil Journal of Phytopathology 173 (3), e70061 , 2025 2025 Citations: 1
A detection of amblyopia using image processing and machine learning techniques K Praveen, S Lalitha, S Gopinath J. Comput. Sci. Eng. Softw. Test. 5 (2), 1-11 , 2023 2023 Citations: 1
A detection of amblyopia medical condition in biomedical datasets using image segmentation and detection processing S Lalitha, N Shanthi, S Gopinath Journal of Medical Imaging and Health Informatics 11 (11), 2814-2821 , 2021 2021 Citations: 1
A PRIVACY-PRESERVING ACCESS CONTROL WITH ROBUST DATA AUTHENTICITY FOR CLOUD GROUP P Sathya, S Gopinath, P Kuppusamy IJETCSE 19 (1), 35-40 , 2015 2015 Citations: 1
Machine Learning Model Based Urban Temperature Analysis with Fuzzy Reinforcement Neural Network. L Pallavi, G Shravani, JS Devi, BS Lakshmi, M Pushpalatha, S Gopinath, ... Journal of Intelligent Systems & Internet of Things 18 (1) , 2026 2026
Image Processing Based Automatic Traffic Control System S Ponnuru, AK Sharma, PF Khan, E Poovannan, BV Lakshmi, S Gopinath 2024 International Conference on Innovative Computing, Intelligent … , 2024 2024
Auguring and Predictions of COVID-19 by ML and PowerBI BG Geetha, K Sakthivel, S Gopinath, G Kowsalya 2023 9th International Conference on Smart Structures and Systems (ICSSS), 1-4 , 2023 2023
Automated Vision Defect Detection Supported Deep Convolutional Neural Networks S Lalitha, N Shanthi, S Gopinath Journal of Physics: Conference Series 1964 (4), 042044 , 2021 2021
LOGICAL MODEL TO DISCOVER SPITEFUL NODES IN MANETS gopinath s International Journal of Current Trends in Engineering & Research 2 (5), 676-684 , 2016 2016
Adaption for Improved Lifetime in Wireless Sensor Networks with Graph Factorization G S International Journal Of Infinite Innovations In Engineering And Technology … , 2015 2015
Privacy Preserving and Truthful Detection of Packet Dropping Attacks in Wireless Adhoc Networks G S International Journal of Science and Engineering Research 3 (10) , 2015 2015
Geographic Routing Based Adaptive Location Update for Mobile Adhoc Networks International Journal Of Infinite Innovations In Engineering And Technology … , 2015 2015
