Scholarlink

<p>Tomato farming is a vital agricultural practice globally but is frequently affected by a range of diseases caused by biotic factors such as fungi, bacteria, and viruses, as well as abiotic stresses including variations in soil pH, moisture levels, temperature, and other environmental conditions. Traditional methods for diagnosing these diseases are often time-consuming, require specialized expertise, and can be expensive, limiting their practical application in large-scale farming. This study proposes the use of advanced deep learning architectures, specifically VGG19 and InceptionV3, to accurately detect and classify various tomato diseases by incorporating both biotic and abiotic factors. The approach involves collecting a diverse dataset, applying systematic preprocessing and feature extraction, and addressing class imbalance to improve model robustness. Experimental results demonstrate that the proposed models achieve a testing accuracy of up to 86.38%, indicating strong potential for real-world deployment in automated disease diagnosis. This work contributes to precision agriculture by enabling early detection and management of tomato diseases, thereby helping to increase crop yield and reduce economic losses.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Deep Learning-Driven Framework for Tomato Disease Analysis: A Biotic and Abiotic Perspective with VGG19 and Inception

  • 期刊论文

  • 发表日期：

    2025年9月24日

查看全文
  • H Najmusher,
  • K Meena

摘要

Tomato farming is a vital agricultural practice globally but is frequently affected by a range of diseases caused by biotic factors such as fungi, bacteria, and viruses, as well as abiotic stresses including variations in soil pH, moisture levels, temperature, and other environmental conditions. Traditional methods for diagnosing these diseases are often time-consuming, require specialized expertise, and can be expensive, limiting their practical application in large-scale farming. This study proposes the use of advanced deep learning architectures, specifically VGG19 and InceptionV3, to accurately detect and classify various tomato diseases by incorporating both biotic and abiotic factors. The approach involves collecting a diverse dataset, applying systematic preprocessing and feature extraction, and addressing class imbalance to improve model robustness. Experimental results demonstrate that the proposed models achieve a testing accuracy of up to 86.38%, indicating strong potential for real-world deployment in automated disease diagnosis. This work contributes to precision agriculture by enabling early detection and management of tomato diseases, thereby helping to increase crop yield and reduce economic losses.