To Classify The Detection of Plant Diseases With Organic Solution Using Deep Learning-Based Approach

The increasing prevalence of plant diseases poses a significant threat to agricultural productivity, food security, and sustainable crop production. Conventional disease management practices frequently rely on synthetic pesticides, whose excessive use can adversely affect soil health, water quality, biodiversity, and human health. Organic treatment solutions provide a more environmentally sustainable alternative; however, their integration with automated plant disease detection and post-treatment monitoring remains relatively underexplored. This study proposes an image-based deep-learning framework for plant disease classification and organic treatment assessment using leaf images. A comprehensive dataset comprising healthy and diseased leaf images was developed using publicly available datasets and field-collected samples, with additional organically treated samples incorporated to represent post-treatment conditions. The proposed framework employs transfer learning with an EfficientNet-B4 based architecture enhanced with attention mechanisms and feature-level integration of color, texture, and shape characteristics. The model was trained and evaluated using accuracy, precision, recall, F1-score, ROC-AUC, and confusion matrix analysis. Experimental results demonstrate that the proposed framework achieved a reported overall accuracy of 96.72%, with precision, recall, F1-score, and ROC-AUC values of 0.963, 0.968, 0.965, and 0.982 respectively. Comparative evaluation demonstrated that the proposed framework outperformed SVM, ResNet50, and conventional EfficientNet-B4 baseline models. The proposed system further provides disease-specific organic treatment recommendations through a knowledge-based decision module. The framework has potential for integration into mobile and edge-based agricultural applications, enabling rapid disease diagnosis, post-treatment monitoring, and environmentally responsible crop management.

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-18
DOI
https://doi.org/10.5281/zenodo.24337111
Primary Topic
Smart Agriculture and AI
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

To Classify The Detection of Plant Diseases With Organic Solution Using Deep Learning-Based Approach

Mohit Sharma, Pradeep
6 links
Zenodo (CERN European Organization for Nuclear Research)
Smart Agriculture and AI
article

To Classify The Detection of Plant Diseases With Organic Solution Using Deep Learning-Based Approach

Mohit Sharma, Pradeep
article en

Abstract

The increasing prevalence of plant diseases poses a significant threat to agricultural productivity, food security, and sustainable crop production. Conventional disease management practices frequently rely on synthetic pesticides, whose excessive use can adversely affect soil health, water quality, biodiversity, and human health. Organic treatment solutions provide a more environmentally sustainable alternative; however, their integration with automated plant disease detection and post-treatment monitoring remains relatively underexplored. This study proposes an image-based deep-learning framework for plant disease classification and organic treatment assessment using leaf images. A comprehensive dataset comprising healthy and diseased leaf images was developed using publicly available datasets and field-collected samples, with additional organically treated samples incorporated to represent post-treatment conditions. The proposed framework employs transfer learning with an EfficientNet-B4 based architecture enhanced with attention mechanisms and feature-level integration of color, texture, and shape characteristics. The model was trained and evaluated using accuracy, precision, recall, F1-score, ROC-AUC, and confusion matrix analysis. Experimental results demonstrate that the proposed framework achieved a reported overall accuracy of 96.72%, with precision, recall, F1-score, and ROC-AUC values of 0.963, 0.968, 0.965, and 0.982 respectively. Comparative evaluation demonstrated that the proposed framework outperformed SVM, ResNet50, and conventional EfficientNet-B4 baseline models. The proposed system further provides disease-specific organic treatment recommendations through a knowledge-based decision module. The framework has potential for integration into mobile and edge-based agricultural applications, enabling rapid disease diagnosis, post-treatment monitoring, and environmentally responsible crop management.

Zenodo (CERN European Organization for Nuclear Research)
Zero hunger
Smart Agriculture and AI
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon