Scholarlink

Finding plant disease is essential for food availability, sustainable farming methods, and crop health and productivity. The tool offers real-time intervention and resulting losses, and resource utilization benefits for agriculture. This paper represents a forward step compared to the image analysis techniques used in studying plant diseases or trait types. In our article, we survey and examine previous studies on leaf disease detection. Additionally, a model that works well and will advance the readers’ work is proposed. Data balancing plays an important role in the proper training of classifying models. Here data augmentation is employed to augment the dataset's extent and ensure its balance. A model containing different convolutional layers is proposed for leaf disease detection from the augmented images. The RmsProp optimization algorithm is used to minimize the error while training the model with image dataset. To assess the feasibility of the proposed model, a widely accessible PlantVillage dataset is utilized to train, validate, and test the proposed deep learning network. The proposed approach provided 99.29% testing accuracy.

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

Plant Disease Detection from Leaf Images Using Augmentation and Convolutional Neural Network

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Pranita Mohapatro,
  • Hitesh Mohapatra,
  • Soumya Ranjan Mishra,
  • Abhishek Das

摘要

Finding plant disease is essential for food availability, sustainable farming methods, and crop health and productivity. The tool offers real-time intervention and resulting losses, and resource utilization benefits for agriculture. This paper represents a forward step compared to the image analysis techniques used in studying plant diseases or trait types. In our article, we survey and examine previous studies on leaf disease detection. Additionally, a model that works well and will advance the readers’ work is proposed. Data balancing plays an important role in the proper training of classifying models. Here data augmentation is employed to augment the dataset's extent and ensure its balance. A model containing different convolutional layers is proposed for leaf disease detection from the augmented images. The RmsProp optimization algorithm is used to minimize the error while training the model with image dataset. To assess the feasibility of the proposed model, a widely accessible PlantVillage dataset is utilized to train, validate, and test the proposed deep learning network. The proposed approach provided 99.29% testing accuracy.