Scholarlink

India’s economy is supported by agriculture crops, which is one of the reasons why it is considered to be one of the developing nations in the globe. The detection of crop diseases is still a challenging process, despite the fact that crop diseases pose a significant threat to the nation’s food supply. ANN, SVM, k-NN, and CNN are only some of the classic classification methods that can be utilized in this context. Both Machine Learning and Deep Learning are also available for application. As a consequence of this, the accuracy of such systems has reached its maximum level because they are dependent on a feature extraction process that is built by hand. For the purpose of improving the accuracy of our classification method, it is essential to merge more than one deep learning model for the classification of maize leaf disease. Therefore, in order to effectively address this matter, we have developed a hybrid plant disease recognition model in order to enhance the learning outcomes of Deep CNN. This was accomplished by combining two models, namely VGG-16 and optimized CNN model, into a single model. For the goal of evaluating the effectiveness of the maize crop in the experiments, 4,400 images of maize crop leaves were taken from the Village Net dataset. These images included Common_rust, Gray_leaf_spot, Healthy and Northern_leaf_blight. According to the findings of our investigation, we have shown that a hybrid maize disease recognition model (VGG-16 Net model and Optimized CNN model) performs better than separate deep learning models (VGG-16 Net model and Optimized CNN model).

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

Integrating Optimized CNN and Deep CNN Model for Enhanced Maize Plant Leaf Disease Classification and Prediction Systems

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • P. Deepan,
  • R. Vidya,
  • K. Kalaivani,
  • J. Ganesh,
  • S. Dhiravidaselvi

摘要

India’s economy is supported by agriculture crops, which is one of the reasons why it is considered to be one of the developing nations in the globe. The detection of crop diseases is still a challenging process, despite the fact that crop diseases pose a significant threat to the nation’s food supply. ANN, SVM, k-NN, and CNN are only some of the classic classification methods that can be utilized in this context. Both Machine Learning and Deep Learning are also available for application. As a consequence of this, the accuracy of such systems has reached its maximum level because they are dependent on a feature extraction process that is built by hand. For the purpose of improving the accuracy of our classification method, it is essential to merge more than one deep learning model for the classification of maize leaf disease. Therefore, in order to effectively address this matter, we have developed a hybrid plant disease recognition model in order to enhance the learning outcomes of Deep CNN. This was accomplished by combining two models, namely VGG-16 and optimized CNN model, into a single model. For the goal of evaluating the effectiveness of the maize crop in the experiments, 4,400 images of maize crop leaves were taken from the Village Net dataset. These images included Common_rust, Gray_leaf_spot, Healthy and Northern_leaf_blight. According to the findings of our investigation, we have shown that a hybrid maize disease recognition model (VGG-16 Net model and Optimized CNN model) performs better than separate deep learning models (VGG-16 Net model and Optimized CNN model).