Authors :
Wijesekara JPD; Waruna Henarangoda; Pavithra Subhashini
Volume/Issue :
Volume 9 - 2024, Issue 1 - January
Google Scholar :
http://tinyurl.com/4cc4uv5j
Scribd :
http://tinyurl.com/34wecm6h
DOI :
https://doi.org/10.5281/zenodo.10609342
Abstract :
The integration of smart farming systems
and requisite infrastructural developments represents a
paradigm shift in agricultural technology, significantly
augmenting both the quality and yield within the sector.
Tomatoes, as one of the world's most vital crops, are
frequently afflicted by leaf diseases, which critically
impact harvest outcomes. Prompt detection and
identification of these diseases are imperative to
mitigate crop devastation and implement efficacious
control measures, particularly in understanding the
pathogen species composition. Delays in disease
diagnosis and inadequate control responses can
precipitate substantial crop losses and marked
degradation in product quality. This study introduces
an IT-based solution leveraging image processing and
deep learning methodologies for the expedited detection
of diseases in tomato plants. Utilizing a dataset of 22,930
images, encompassing nine distinct diseased-leaf
categories and a healthy-leaf category, the research
employs a Convolutional Neural Network (CNN) for
disease classification and prediction. The model
demonstrates notable efficacy, achieving an overall
accuracy rate of 98.2% and maintaining a loss rate of
0.0532. This advancement in precision agriculture
exemplifies the potential of integrating cutting-edge
technology with traditional farming practices to
enhance productivity and disease management.
Keywords :
Component, formatting, style, styling, insert.
The integration of smart farming systems
and requisite infrastructural developments represents a
paradigm shift in agricultural technology, significantly
augmenting both the quality and yield within the sector.
Tomatoes, as one of the world's most vital crops, are
frequently afflicted by leaf diseases, which critically
impact harvest outcomes. Prompt detection and
identification of these diseases are imperative to
mitigate crop devastation and implement efficacious
control measures, particularly in understanding the
pathogen species composition. Delays in disease
diagnosis and inadequate control responses can
precipitate substantial crop losses and marked
degradation in product quality. This study introduces
an IT-based solution leveraging image processing and
deep learning methodologies for the expedited detection
of diseases in tomato plants. Utilizing a dataset of 22,930
images, encompassing nine distinct diseased-leaf
categories and a healthy-leaf category, the research
employs a Convolutional Neural Network (CNN) for
disease classification and prediction. The model
demonstrates notable efficacy, achieving an overall
accuracy rate of 98.2% and maintaining a loss rate of
0.0532. This advancement in precision agriculture
exemplifies the potential of integrating cutting-edge
technology with traditional farming practices to
enhance productivity and disease management.
Keywords :
Component, formatting, style, styling, insert.
