Semantic segmentation based on enhanced gated pyramid network with lightweight attention module A. Viswanathan, V. Senthil kumar, M. Umamaheswari, Vignesh Janarthanan, M. Jaganathan AI Communications, 2024 Semantic segmentation has made tremendous progress in recent years. The development of large datasets and the regression of convolutional models have enabled effective training of very large semantic model. Nevertheless, higher capacity indicates a higher computational problem, thus preventing real-time operation. Yet, due to the limited annotations, the models may have relied heavily on the available contexts in the training data, resulting in poor generalization to previously unseen scenes. Therefore, to resolve these issues, Enhanced Gated Pyramid network (GPNet) with Lightweight Attention Module (LAM) is proposed in this paper. GPNet is used for semantic feature extraction and GPNet is enhanced by the pre-trained dilated DetNet and Dense Connection Block (DCB). LAM approach is applied to habitually rescale the different feature channels weights. LAM module can increase the accuracy and effectiveness of the proposed methodology. The performance of proposed method is validated using Google Colab environment with different datasets such as Cityscapes, CamVid and ADE20K. The experimental results are compared with various methods like GPNet-ResNet-101 and GPNet-ResNet-50 in terms of IoU, precision, accuracy, F1 score and recall. From the overall analysis cityscapes dataset achieves 94.82% pixel accuracy.
Rice leaf disease detection based on bidirectional feature attention pyramid network with YOLO v5 model V Senthil Kumar, M Jaganathan, A Viswanathan, M Umamaheswari, J Vignesh Environmental Research Communications, 2023 To ensure higher quality, capacity, and production of rice, it is vital to diagnose rice leaf disease in its early stage in order to decrease the usage of pesticides in agriculture which in turn avoids environmental damage. Hence, this article presents a Multi-scale YOLO v5 detection network to detect and classify the rice crop disease in its early stage. The experiment is initially started by pre-processing the rice leaf images obtained from the RLD dataset, after which data set labels are created, which are then divided into train and test sets. DenseNet-201 is used as the backbone network and depth-aware instance segmentation is used to segment the different regions of rice leaf. Moreover, the proposed Bidirectional Feature Attention Pyramid Network (Bi-FAPN) is used for extracting the features from the segmented image and also enhances the detection of diseases with different scales. Furthermore, the feature maps are identified in the detection head, where the anchor boxes are then applied to the output feature maps to produce the final output vectors by the YOLO v5 network. The subset of channels or filters is pruned from different layers of deep neural network models through the principled pruning approach without affecting the full framework performance. The experiments are conducted with RLD dataset with different existing networks to verify the generalization ability of the proposed model. The effectiveness of the network is evaluated based on various parameters in terms of average precision, accuracy, average recall, IoU, inference time, and F1 score, which are achieved at 82.8, 94.87, 75.81, 0.71, 0.017, and 92.45 respectively.
Creating an ML-based Mobile App for Maintaining Crop Health A. Viswanathan, M. Umamaheswari, M. Jaganathan, Julius B. Wosowei, Rohi Prasad, P. Indira International Conference on Edge Computing and Applications Icecaa 2022 Proceedings, 2022 Agriculture is the backbone of our nation. Not only in India, but the growth of food and other crops is also an essential occupation all over the world. One of the major requirements of great yield in agriculture is the analysis of the soil and the atmosphere. This study aims in finding the best Machine Learning (ML) model that can be used to determine the right fertilizer based on all the factors like environmental conditions and crop requirements. For this purpose, three different ML models were developed using three different algorithms. The algorithms used in this study were the Neural Network (NN), Classification and Regression Tree (CART), and Linear Discriminant Analysis (LDA). Before the models were constructed, a dataset consisting of all necessary data about crops and the environment is obtained from Kaggle. This dataset then undergoes a couple of processes named encoding and upsampling to ensure the better performance of the ML models. The models are then trained using this preprocessed dataset. An unused part of the dataset is then used to test the efficiency of the models. This analysis is done using four parameters named the True Positive Rate (TPR), True Negative Rate (TNR), False Positive Rate (FPR), and False Negative Rate (FNR). By the end of this analysis, it is found that the NN algorithm is the best algorithm that can be used to predict the best fertilizer with the perfect score of 100 as a true negative rate. The LDA algorithm has the least right predictions and the maximum number of false predictions making it the worst algorithm among the three. The best algorithm, i.e., the model designed using the NN algorithm is deployed into a software application. Users can enter all the necessary details to find the most suitable fertilizer for crops based on environmental factors and crop requirements.
Parallel heuristic based segmentation technique Journal of Advanced Research in Dynamical and Control Systems, 2017
RECENT SCHOLAR PUBLICATIONS
A multi-modal ensemble framework for breast cancer segmentation and classification using genetic U-Net and HBoost KV Guru, V Janarthanan, M Jaganathan Biomedical Signal Processing and Control 112, 108827 , 2026 2026.0 Citations: 1
Semantic segmentation based on enhanced gated pyramid network with lightweight attention module A Viswanathan, VS Kumar, M Umamaheswari, V Janarthanan, ... AI Communications 37 (1), 97-114 , 2024 2024.0 Citations: 1
Rice leaf disease detection based on bidirectional feature attention pyramid network with YOLO v5 model VS Kumar, M Jaganathan, A Viswanathan, M Umamaheswari, J Vignesh Environmental Research Communications 5 (6), 065014 , 2023 2023.0 Citations: 86
Creating an ML-based Mobile App for Maintaining Crop Health A Viswanathan, M Umamaheswari, M Jaganathan, JB Wosowei, ... 2022 International Conference on Edge Computing and Applications (ICECAA … , 2022 2022.0 Citations: 2
Minimizing the Misinformation in Social Networksusing Heuristic Greedy Algorithm M Jaganathan, A Kumar, NK Keerthana, S Teja 2022.0
Smart Home Using IoT JM Elango M, Gokulprasanth C, Heyram K International Journal for Science and Advance Research In Technology 7 (4 … , 2021 2021.0
An heuristic cloud based segmentation technique using edge and texture based two dimensional entropy M Jaganathan, A Sabari Cluster Computing 22 (Suppl 5), 12767-12776 , 2019 2019.0 Citations: 18
Modelling an effectual feature selection approach for predicting down syndrome using machine learning approaches M Jaganathan, R Gopal, VR Kiruthika International Journal of Aquatic Science, 1238-1249 , 2019 2019.0 Citations: 6
PARALLEL HEURISTIC BASED SEGMENTATION TECHNIQUE DAS Jaganathan M Journal of Advanced research in dynamical and Control systems, 2480-2493 , 2017 2017.0
Content Protection System Using Matching Object for Cloud Based Multimedia. JM Gayathri S, Priyanka M International Journal on Applications in Information and Communication … , 2016 2016.0
Vehicular Tracking and Reporting by Vehicle to Vehicle (V2V) Communication MJ A.B.Yuvaraj International Journal of Innovative Research in Science, Engineering and … , 2015 2015.0
STRESS DETECTION IN IT PROFESSIONALS M Meghana, RL Mahitha, N Deepika, V Sneha, M Jaganathan
MOST CITED SCHOLAR PUBLICATIONS
Rice leaf disease detection based on bidirectional feature attention pyramid network with YOLO v5 model VS Kumar, M Jaganathan, A Viswanathan, M Umamaheswari, J Vignesh Environmental Research Communications 5 (6), 065014 , 2023 2023.0 Citations: 86
An heuristic cloud based segmentation technique using edge and texture based two dimensional entropy M Jaganathan, A Sabari Cluster Computing 22 (Suppl 5), 12767-12776 , 2019 2019.0 Citations: 18
Modelling an effectual feature selection approach for predicting down syndrome using machine learning approaches M Jaganathan, R Gopal, VR Kiruthika International Journal of Aquatic Science, 1238-1249 , 2019 2019.0 Citations: 6
Creating an ML-based Mobile App for Maintaining Crop Health A Viswanathan, M Umamaheswari, M Jaganathan, JB Wosowei, ... 2022 International Conference on Edge Computing and Applications (ICECAA … , 2022 2022.0 Citations: 2
A multi-modal ensemble framework for breast cancer segmentation and classification using genetic U-Net and HBoost KV Guru, V Janarthanan, M Jaganathan Biomedical Signal Processing and Control 112, 108827 , 2026 2026.0 Citations: 1
Semantic segmentation based on enhanced gated pyramid network with lightweight attention module A Viswanathan, VS Kumar, M Umamaheswari, V Janarthanan, ... AI Communications 37 (1), 97-114 , 2024 2024.0 Citations: 1
Minimizing the Misinformation in Social Networksusing Heuristic Greedy Algorithm M Jaganathan, A Kumar, NK Keerthana, S Teja 2022.0
Smart Home Using IoT JM Elango M, Gokulprasanth C, Heyram K International Journal for Science and Advance Research In Technology 7 (4 … , 2021 2021.0
PARALLEL HEURISTIC BASED SEGMENTATION TECHNIQUE DAS Jaganathan M Journal of Advanced research in dynamical and Control systems, 2480-2493 , 2017 2017.0
Content Protection System Using Matching Object for Cloud Based Multimedia. JM Gayathri S, Priyanka M International Journal on Applications in Information and Communication … , 2016 2016.0
Vehicular Tracking and Reporting by Vehicle to Vehicle (V2V) Communication MJ A.B.Yuvaraj International Journal of Innovative Research in Science, Engineering and … , 2015 2015.0
STRESS DETECTION IN IT PROFESSIONALS M Meghana, RL Mahitha, N Deepika, V Sneha, M Jaganathan
