深層学習によるカンキツグリーニング病の簡易診断技術の開発

カテゴリ: 論文誌(論文単位)

グループ名: 【C】電子・情報・システム部門

発行日: 2024/08/01

タイトル(英語): A Simple Diagnostic Method for Citrus Greening Disease with Deep Learning

著者名: Ruihao Dong（関西大学），白岩　史（鳥取大学），林　武文（関西大学）

著者名(英語): Ruihao Dong (Kansai University), Aya Shiraiwa (Tottori University), Takefumi Hayashi (Kansai University)

キーワード: 目標検出，植物病害検出，カンキツグリーニング病，Faster RCNN，転移学習　　object detection，plant disease detection，citrus greening disease，Faster RCNN，transfer learning

要約(英語): Citrus Greening disease (CG) is the most destructive disease of citrus, leading to branch dieback and plant death. Currently, there is no cure for CG, the early detection and removal of infected trees is important to prevent the spread of the disease. In recent years, there have been growing expectations for CG detection with digital images, especially deep learning techniques applied to digitized herbarium specimen image data. However, this approach faces challenges in practical applicability and detection efficiency. In this paper, we proposed a simple diagnostic method for CG using transfer learning with the Faster RCNN architecture. We collected in-field images from a citrus orchard in Thailand where CG has caused significant damage. We compared the performance of two annotation methods with the in-field leaf dataset and discussed their effects on pre-trained VGG and Resnet models. 5-fold cross-validation was utilized for model training and evaluation, with Average Precision (AP) used as the performance metric. The results showed that the Resnet models performed better than the VGG models, with the Resnet152 model scoring the highest in this task. The annotation method which including annotations of healthy and other diseases leaves achieved an AP of 84.13% lower than another one but indicated better performance in practical applications with more robustness. Additionally, we developed a web application that performs real-time diagnosis using our trained models and verified the effectiveness of our system.

本誌: 電気学会論文誌C（電子・情報・システム部門誌） Vol.144 No.8 （2024） 特集Ⅰ：Smart Cityを支える高度な無線通信，センシング及び情報処理　特集Ⅱ：人と人とを繋ぐ情報・システム技術

本誌掲載ページ: 824-830 p

原稿種別: 論文／日本語

電子版へのリンク: https://www.jstage.jst.go.jp/article/ieejeiss/144/8/144_824/_article/-char/ja/