Scholarlink

<p>This research presents a new combined deep learning system for effective and reliable identification of plant diseases in complicated agricultural environments. One of the most difficult jobs in agriculture is identifying plant diseases early on. Early disease detection in plants is crucial for increasing agricultural yield. With the application of machine learning and deep learning techniques, this issue has been resolved. Large crop farms can now detect plant illnesses automatically, which is advantageous as it reduces the monitoring time. The suggested approach consists of multiple important stages. To begin with, image quality of the agricultural lands is improved through preprocessing techniques like noise reduction, gamma correction and white balancing. Data augmentation is incorporated to expand the dataset and improve the generalization capacity of the model. Efficient methods such as EfficientDet, Squeeze Net, colour and shape based features, are included in feature extraction. The most relevant features are selected by a Hybrid Optimization Algorithm (HOA), which integrates Mother Optimization Algorithm (MOA), Teaching learning-based optimization (TLBO) and Improved Wild Horse Optimization (IWHO) to detect the various plant diseases like Bacterial Blight, Tungro, Blast and Brown spot. At last, a deep learning detector, which may include Recurrent Convolutional Neural Networks (RCNN) and Recurrent Neural Networks (RNN), Deep Neural Networks (DNN) and Bidirectional Long Short-Term Memory (Bi-LSTM) identifies the location and type of plant diseases. Finally, HOA integrated with MOA, TLBO, IWHO feature selection and CNN classifier is proposed for identification of planst diseases from the hyper spectral images. Gradient Descent (GD) method is used for weight adjustment to tune the hyper parameters so as to reduce the loss function. It is also implemented to avoid over fitting and improve the overall generalization. This comprehensive approach depicts encouraging results in overcoming challenges in plant disease detection.</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

Hybrid deep learning approach for early detection and classification of plant diseases from aerial view of agricultural lands

  • 期刊论文

  • 发表日期：

    2025年6月28日

查看全文
  • K. Sujatha,
  • N. P. G. Bhavani,
  • G. Rohini,
  • V. Srividhya,
  • A. Ganesan

摘要

This research presents a new combined deep learning system for effective and reliable identification of plant diseases in complicated agricultural environments. One of the most difficult jobs in agriculture is identifying plant diseases early on. Early disease detection in plants is crucial for increasing agricultural yield. With the application of machine learning and deep learning techniques, this issue has been resolved. Large crop farms can now detect plant illnesses automatically, which is advantageous as it reduces the monitoring time. The suggested approach consists of multiple important stages. To begin with, image quality of the agricultural lands is improved through preprocessing techniques like noise reduction, gamma correction and white balancing. Data augmentation is incorporated to expand the dataset and improve the generalization capacity of the model. Efficient methods such as EfficientDet, Squeeze Net, colour and shape based features, are included in feature extraction. The most relevant features are selected by a Hybrid Optimization Algorithm (HOA), which integrates Mother Optimization Algorithm (MOA), Teaching learning-based optimization (TLBO) and Improved Wild Horse Optimization (IWHO) to detect the various plant diseases like Bacterial Blight, Tungro, Blast and Brown spot. At last, a deep learning detector, which may include Recurrent Convolutional Neural Networks (RCNN) and Recurrent Neural Networks (RNN), Deep Neural Networks (DNN) and Bidirectional Long Short-Term Memory (Bi-LSTM) identifies the location and type of plant diseases. Finally, HOA integrated with MOA, TLBO, IWHO feature selection and CNN classifier is proposed for identification of planst diseases from the hyper spectral images. Gradient Descent (GD) method is used for weight adjustment to tune the hyper parameters so as to reduce the loss function. It is also implemented to avoid over fitting and improve the overall generalization. This comprehensive approach depicts encouraging results in overcoming challenges in plant disease detection.