Multiple disease detection method for greenhouse-cultivated strawberry based on multiscale feature fusion Faster R_CNN

Abstract

Disease has a significant impact on strawberry quality and yield, and deep learning has become an important approach for the detection of crop disease. To address the problems of complex backgrounds and small disease spots in strawberry disease images from natural environments, we propose a new Faster R_CNN architecture. The multiscale feature fusion network is composed of ResNet, FPN, and CBAM blocks, and it can effectively extract rich strawberry disease features. We built a dataset for strawberry leaves, flowers and fruits, and the experimental results showed that the model was able to effectively detect healthy strawberries and seven strawberry diseases under natural conditions, with an mAP of 92.18% and an average detection time of only 229 ms. The model is compared with Mask R_CNN and YOLO-v3, and we find that our model can guarantee high accuracy and fast detection operational requirements. Our method provides an effective solution for crop disease detection and can improve farmers' management of the strawberry growing process.

Plant phenotyping relevance

イチゴの葉・花・果実画像から病害状態を検出する深層学習手法を開発し、データセット構築、比較評価、精度・速度検証を行っており、植物フェノタイピング手法が研究の中心である。

Code and data availability

公開本文の所在を確認できませんでした。非公開または購読が必要な可能性があります。

No evidence-backed public reproduction asset is currently recorded.