Scholarlink

Early detection and reliable classification of tomato leaf diseases are critical to reducing yield losses and enabling targeted interventions in precision agriculture. This paper presents an end-to-end pipeline coupling real-time object detection for tomato leaves with image classification of foliar diseases. We compare YOLOv11n and RT-DETR for leaf localization under varied backgrounds and benchmark CNN classifiers (MobileNetV3, ResNet50, EfficientNetB5) for disease recognition. The final pipeline integrates ONNX/TensorRT for embedded deployment on Jetson Nano and achieves competitive detection mAP and classification F1-score.

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

Intelligent System for Detection and Classification of Tomato Leaf Diseases

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Asmae Salih,
  • El Mehdi Cherrat,
  • Azzedine Dliou,
  • Amine Saddik,
  • Mohamed Zarboubi

摘要

Early detection and reliable classification of tomato leaf diseases are critical to reducing yield losses and enabling targeted interventions in precision agriculture. This paper presents an end-to-end pipeline coupling real-time object detection for tomato leaves with image classification of foliar diseases. We compare YOLOv11n and RT-DETR for leaf localization under varied backgrounds and benchmark CNN classifiers (MobileNetV3, ResNet50, EfficientNetB5) for disease recognition. The final pipeline integrates ONNX/TensorRT for embedded deployment on Jetson Nano and achieves competitive detection mAP and classification F1-score.