DSpace Repository

Sunflower Disease Identification using Deep Learning: A data-driven approach

Show simple item record

dc.contributor.author Sohel, Amir
dc.contributor.author Sarker, Md. Murshidul Alam
dc.contributor.author Das, Utpal Chandra
dc.contributor.author Das, Pranajit Kumar
dc.contributor.author Siddiquee, Shah Md Tanvir
dc.contributor.author Noori, Sheak Rashed Haider
dc.contributor.author Le, Ngoc Thien
dc.date.accessioned 2024-08-29T06:39:53Z
dc.date.available 2024-08-29T06:39:53Z
dc.date.issued 2024-04-12
dc.identifier.uri http://dspace.daffodilvarsity.edu.bd:8080/handle/123456789/13285
dc.description.abstract The sunflower (Helianthus annuus) is considered to possess a low to moderate susceptibility to drought conditions. However, production has reduced as a result of some of its illnesses. Therefore, it is necessary to take steps to identify the diseases of plants earlier. Disease identification is a time-consuming process that cannot be achieved simultaneously. In this study we have proposed a novel strategy utilizing deep learning techniques to precisely detect and classify three distinct diseases affecting sunflower leaves, as well as molds, based on image analysis. A collection of 466 images depicting four distinct types, namely Gray mold, downy mildew, leaf scars, and healthy leaves, were gathered from sunflower leaves and molds found in the agricultural fields of Bangladesh. The dataset undergoes several images preprocessing techniques, including rescaling, augmentation, histogram equalization, contrast stretching, gamma correction, Gaussian noise, and Gaussian filtering. Subsequently, feature extraction methodologies are employed for figuring out the underlying cause of infection in a picture. Four separate evaluation metrics have been implemented in order to assess the performance of each classifier. We have employed a total of five deep-learning methodologies named as InceptionV3, VGG19, MobileNetV2, ResNet152V2, DenseNet201. Among the models taken into consideration, DenseNet201 had the most optimal performance, with a notable accuracy rate of 98.7%. © 2023 IEEE. en_US
dc.language.iso en_US en_US
dc.publisher IEEE en_US
dc.subject Deep learning en_US
dc.subject Plant disease en_US
dc.title Sunflower Disease Identification using Deep Learning: A data-driven approach en_US
dc.type Article en_US


Files in this item

This item appears in the following Collection(s)

Show simple item record

Search DSpace


Browse

My Account