Scholarlink

Alternaria Solani is a fungus that causes blight disease in potatoes and tomatoes. The illness mostly affects older plant tissue, resulting in collapse and decay. Each crop of potatoes (Solanum tuberosum L.) and tomatoes (Solanum lycopersicum L.) is frequently affected by two diseases namely early stage foliage and delayed fungal diseases. Utilizing a multi-layered model based Deep Residual Network-9 model, the blight illness has been identified, allowing farmers to take advantage of the disease’s current state. In order to forecast outcomes, the residual Network-9 model takes into account both the healthy and afflicted leaf forms. More pre-trained models that are intended to better handle this problem are EfficientNet, 16-Layer Deep Convolutional Network, InceptionV3, and AlexNet, which extract features on many scales using varying kernel sizes. We intend to compare the models’ increased accuracies by putting these models into practice. The proposed deep learning technique will be trained and verified on a dataset containing leaves from tomato (Solanum lycopersicum L.) and potato (Solanum tuberosum L.) plants.

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

Deep Learning Models for Blight Threat Detection in Potato and Tomato: A Comparative Evaluation

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • R. Devika,
  • S. Dharuna,
  • M. Harini,
  • V. S. Sowbarani

摘要

Alternaria Solani is a fungus that causes blight disease in potatoes and tomatoes. The illness mostly affects older plant tissue, resulting in collapse and decay. Each crop of potatoes (Solanum tuberosum L.) and tomatoes (Solanum lycopersicum L.) is frequently affected by two diseases namely early stage foliage and delayed fungal diseases. Utilizing a multi-layered model based Deep Residual Network-9 model, the blight illness has been identified, allowing farmers to take advantage of the disease’s current state. In order to forecast outcomes, the residual Network-9 model takes into account both the healthy and afflicted leaf forms. More pre-trained models that are intended to better handle this problem are EfficientNet, 16-Layer Deep Convolutional Network, InceptionV3, and AlexNet, which extract features on many scales using varying kernel sizes. We intend to compare the models’ increased accuracies by putting these models into practice. The proposed deep learning technique will be trained and verified on a dataset containing leaves from tomato (Solanum lycopersicum L.) and potato (Solanum tuberosum L.) plants.