用三模型集成提升农作物病害识别准确率，达99.23%
AgriMind: An Ensemble Deep Learning Framework for Multi-Class Plant Disease Classification
- 融合ResNet50、EfficientNet-B0与DenseNet121，用冻结主干+微调头部
- 集成后测试准确率达99.23%，较单模型降低三分之二错误率
- 适合农业病害自动化检测，尤其适用于移动端部署优化
在孟加拉国，农作物病害检测仍依赖人工目视，覆盖数百万小农户。我们构建了AgriMind系统，通过集成ResNet50、EfficientNet-B0和DenseNet121三个模型，基于20,638张PlantVillage图像（涵盖15种辣椒、土豆、番茄病害类别）进行训练。采用冻结ImageNet预训练主干的迁移学习策略，仅对分类头进行10轮微调，保持轻量化。各模型单独在测试集上达到96%–97%准确率，而融合其softmax输出的集成模型进一步提升至99.23%，错误率减少约三分之二。尝试偏向表现最优模型的加权平均反而适得其反；移除任一模型均导致性能下降。辣椒与土豆病害分类准确率完美，番茄因十类视觉相似，仍达99.01%。在NVIDIA T4 GPU上，完整集成模型推理速度为53 FPS。是否可实现在移动设备上的实时运行，取决于TensorFlow Lite的优化，目前尚未完成。
原文摘要 · Abstract (English)
Plant disease detection is still largely manual in Bangladesh, where extension workers eyeball leaf samples across millions of smallholdings. We built AgriMind to automate this: an ensemble of ResNet50, EfficientNet-B0, and DenseNet121 trained on 20,638 PlantVillage images across 15 pepper, potato, and tomato disease classes. Transfer learning with frozen ImageNet backbones and 10 epochs of head-only training keeps the pipeline lightweight. Individual models hit 96--97% on the held-out test set, but averaging their softmax outputs pushes the ensemble to 99.23% -- a two-thirds cut in error rate. We tried biasing the average toward the best validation model; it backfired. Dropping any single model also hurt. Pepper and potato classify perfectly; tomato, with ten visually similar classes, still reaches 99.01%. On an NVIDIA T4 GPU the full ensemble runs at 53 FPS. Whether that translates to real-time mobile use depends on TensorFlow Lite optimization -- work we have not yet completed.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。