Enhanced-RICAP: A Novel Data Augmentation Strategy for Improved Plant Disease Identification

Strategic Insights -

Researchers from Gansu Agricultural University have introduced a new data augmentation technique called Enhanced-RICAP, designed to improve the accuracy of deep learning models for plant disease detection. This study aimed to address the limitation of traditional augmentation methods by introducing an attention module guided by class activation maps. The method was evaluated using several deep learning architectures on the cassava leaf disease and PlantVillage tomato leaf disease datasets.

Key Takeaways:

  • Enhanced-RICAP is an advanced data augmentation technique that replaces random patch selection with an attention module guided by class activation maps.
  • The method reduces label noise and improves model accuracy for plant disease detection, addressing a key limitation of traditional augmentation methods.
  • Enhanced-RICAP consistently outperforms existing augmentation methods, including CutMix, MixUp, CutOut, Hide-and-Seek, and RICAP, across key evaluation metrics: accuracy, precision, recall, and F1-score.
  • The ResNet18+Enhanced-RICAP configuration achieved 99.86% accuracy on the tomato leaf disease dataset, whereas the Xception+Enhanced-RICAP model attained 96.64% accuracy in classifying four cassava leaf disease categories.
  • To bridge the gap between research and practical application, the ResNet18+Enhanced-RICAP model was deployed in PlantDisease, a mobile application that enables real-time disease identification and management recommendations.
  • Enhanced-RICAP supports sustainable agriculture and strengthens food security by providing farmers with accessible and reliable diagnostic tools.
  • The study was funded by the National Natural Science Foundation of China and the Science and Technology Program of Gansu Province.
  • The research team includes Yue Li, Mamadou Bailo Diallo, Okafor Sylevester Chukwuka, Solomon Boamah, Yuhong Gao, Mohamed Meyer Kana Kone, Gelebo Rocho, and Linjing Wei.

Statistics:

  • 99.86% accuracy achieved by the ResNet18+Enhanced-RICAP configuration on the tomato leaf disease dataset
  • 96.64% accuracy attained by the Xception+Enhanced-RICAP model in classifying four cassava leaf disease categories
  • Enhanced-RICAP outperformed existing augmentation methods, including CutMix, MixUp, CutOut, Hide-and-Seek, and RICAP
  • The study was published in Frontiers in Plant Science, a peer-reviewed scientific journal
  • The research has been funded by the National Natural Science Foundation of China (NSFC) and the Science and Technology Program of Gansu Province

Sources:

  • NewsRx. Researchers at Gansu Agricultural University Detail Findings in Plant Diseases and Conditions (Enhanced-RICAP: a novel data augmentation strategy for improved deep learning-based plant disease identification and mobile diagnosis). Life Science Weekly. October 21, 2025; p 5518.
  • Frontiers in Plant Science. Enhanced-RICAP: a novel data augmentation strategy for improved deep learning-based plant disease identification and mobile diagnosis. 2025;16:1646611.