Scholarlink

Solanum tuberosum also known as Potato is one of the most important crops worldwide, contributing significantly to global food security. However, potato production is slowed down by a variety of diseases that can cause crucial yield losses if not effectively managed. Here, we explore the application of deep learning techniques for the classification of potato diseases based on leaf symptoms, aiming to assist farmers in timely disease diagnosis and management decisions. In this article, we use the convolution neural network (CNN) method; also, examine seven classes of potato diseases with the names: Bacteria, Fungi, Nematode, Phytophthora, Pest, Virus, and Healthy. We used a database of 3076 potato images. This proposed plan achieves an accuracy of 99% with respect to the EfficientNetV2B3 model, MobileNetV3-Large model, VGG-16 model, ResNet50 model, and DenseNet121 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

Potato Disease Classification in an Uncontrolled Environment Using Deep Learning Approaches

  • 章节

  • 发表日期：

    2025年

查看全文
  • Anupam Anubhav,
  • Tanisha Samantaray,
  • Chittaranjan Pradhan

摘要

Solanum tuberosum also known as Potato is one of the most important crops worldwide, contributing significantly to global food security. However, potato production is slowed down by a variety of diseases that can cause crucial yield losses if not effectively managed. Here, we explore the application of deep learning techniques for the classification of potato diseases based on leaf symptoms, aiming to assist farmers in timely disease diagnosis and management decisions. In this article, we use the convolution neural network (CNN) method; also, examine seven classes of potato diseases with the names: Bacteria, Fungi, Nematode, Phytophthora, Pest, Virus, and Healthy. We used a database of 3076 potato images. This proposed plan achieves an accuracy of 99% with respect to the EfficientNetV2B3 model, MobileNetV3-Large model, VGG-16 model, ResNet50 model, and DenseNet121 model.