Browsing by Author "Peyal, H.I."

Now showing 1 - 2 of 2

A Lightweight CNN-SVM Explainable AI Approach for Classification and Visualization of Grape Leaf Disease
(Institute of Electrical and Electronics Engineers Inc., 2024-04-25) Peyal, H.I.; Leion, Z.M.; Abdal, M.N.,; Islam, M.I.; Miraz, S.; Remon, M.R.; Kontho, M.M.R.; Tasnim, N.
Grape is highly esteemed as a significant agricultural crop in Bangladesh. Plant diseases primarily result from the presence of pathogens and pest insects, leading to a significant decline in productivity if not properly addressed. The fundamental aim of this research is to employ a streamlined CNN-SVM architecture, utilizing deep learning techniques, to accurately categorize grape leaves into three distinct disease classes and one healthy class. The proposed model surpasses the accuracy of the previously trained transfer learning models VGG-16 and VGG-19 while having approximately 257× to 267× times fewer parameters (0.537 M). On average, the proposed model achieves a classification accuracy of 99.18%, which is significantly higher than the 93.42% and 91.94% achieved by the transfer learning models, respectively. With a precision, recall, and F1 score close to 99%, the suggested model provides excellent results. The model's outstanding performance is further validated by its remarkable Area Under Curve (AUC) score of 99.98%. In addition to using less disc space (about 6 MB), the suggested model because of being lightweight shows a significant decrease in parameters. To visually display the disease identified by the proposed model, a transparent AI methodology has been utilized, specifically the Gradient Weighted Class Activation Mapping (Grad-CAM) technique. To better understand which area was responsible for the classification, a heatmap has been created.
Detection of Primary Open Angle Glaucoma Based on Deep CNN Using Fundus Images
(Institute of Electrical and Electronics Engineers Inc., 2023-12-13) Ratul, M.T.R.; Afroge, S.; Peyal, H.I.; Zafrin, D.; Ahmed, K.F.; Abdal, M.N.
The early and accurate diagnosis of glaucoma, a primary cause of permanent blindness, is critical for efficient treatment and prevention of vision loss. Although the exact causes of glaucoma are not yet fully understood, it is thought to be a result of several factors, including raised pressure inside the eye and decreased blood supply to the optic nerve. We have developed a convolutional neural network model for accurate detection of glaucoma. Methods based on deep learning have been effective at classifying diseases in retinal fundus images, facilitating in the evaluation of the growing number of images. The goal of this work is to create and train a unique deep CNN model that makes use of the connections between related eye-fundus tasks and metrics used to identify glaucoma. We have meticulously selected two distinct datasets to underpin this research endeavor: the ACRIMA dataset and the LAG dataset. Notably, our model attains a remarkable accuracy score of 99.29% on the ACRIMA dataset and an equally commendable accuracy score of 97.22% on the LAG dataset. This performance eclipses that of the majority of contemporary deep CNN models, underscoring the prowess and sophistication of our approach