Scholarlink

<p>Plant diseases cause significant threat affecting the agricultural production and global food security. Early diagnosis of plant disease is crucial for preventing the spread of plant diseases and improving the production rate efficiently. However, the manual detection of plant diseases is an error-prone and time-consuming process. Recently, the application of advanced technologies such as Machine Learning (ML) and Deep Learning (DL) aids in addressing the challenges, resulting in robust identification of plant diseases. Consequently, this research presents an inclusive analysis of various Deep Learning and Machine Learning models for plant leaf disease classification. In addition, the research includes LBP, LTP, and Wavelet features with ResNet-101 architecture for enhancing the feature representation. Specifically, the proposed method exploits the Dolphin echolocation optimization and Coyote optimization for achieving the optimal tuning of classifiers. Moreover, the study underscores that the DL models achieve exceptional performance compared to ML classifiers. Based on the experimental analysis, the Dolphin echolocation optimization based Deep Convolutional Neural Network (DEO-Deep CNN) model shows superior performance, achieving the highest accuracy of 94.71%, F1-score of 94.80%, sensitivity of 94.35%, and specificity of 95.25% compared to other state-of-the-art methods in plant disease classification with 90% of training on New Plant diseases database. Besides, the research provides insights into potential solutions and outlines the direction for future research in the field of plant disease classification.</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

An efficient plant leaf disease classification using machine learning and deep learning models

  • 期刊论文

  • 发表日期：

    2026年5月8日

查看全文
  • Smita R. Sankhe

摘要

Plant diseases cause significant threat affecting the agricultural production and global food security. Early diagnosis of plant disease is crucial for preventing the spread of plant diseases and improving the production rate efficiently. However, the manual detection of plant diseases is an error-prone and time-consuming process. Recently, the application of advanced technologies such as Machine Learning (ML) and Deep Learning (DL) aids in addressing the challenges, resulting in robust identification of plant diseases. Consequently, this research presents an inclusive analysis of various Deep Learning and Machine Learning models for plant leaf disease classification. In addition, the research includes LBP, LTP, and Wavelet features with ResNet-101 architecture for enhancing the feature representation. Specifically, the proposed method exploits the Dolphin echolocation optimization and Coyote optimization for achieving the optimal tuning of classifiers. Moreover, the study underscores that the DL models achieve exceptional performance compared to ML classifiers. Based on the experimental analysis, the Dolphin echolocation optimization based Deep Convolutional Neural Network (DEO-Deep CNN) model shows superior performance, achieving the highest accuracy of 94.71%, F1-score of 94.80%, sensitivity of 94.35%, and specificity of 95.25% compared to other state-of-the-art methods in plant disease classification with 90% of training on New Plant diseases database. Besides, the research provides insights into potential solutions and outlines the direction for future research in the field of plant disease classification.