Research Article | Open Access

Volume 2025 |Article ID 100024 | https://doi.org/10.1016/j.plaphe.2025.100024

PlantCaFo: An efficient few-shot plant disease recognition method based on foundation models

Xue Jiang,¹ Jiashi Wang,¹ Kai Xie,¹ Chenxi Cui,¹ Aobo Du,¹ Xianglong Shi,¹ Wanneng Yang,² Ruifang Zhai ^1,3,4

¹College of Informatics, Huazhong Agricultural University, Wuhan 430070, PR China
²National Key Laboratory of Crop Genetic Improvement, National Center of Plant Gene Research, Huazhong Agricultural University, Wuhan 430070, PR China
³Engineering Research Center of Intelligent Technology for Agriculture, Ministry of Education, Huazhong Agricultural University, Wuhan 430070, PR China
⁴Hubei Key Laboratory of Agricultural Bioinformatics, College of Informatics, Huazhong Agricultural University, Wuhan 430070, PR China

Abstract

Although plant disease recognition is highly important in agricultural production, traditional methods face challenges due to the high costs associated with data collection and the scarcity of samples. Few-shot plant disease identification tasks, which are based on transfer learning, can learn feature representations from a small amount of data; however, most of these methods require pretraining within the relevant domain. Recently, foundation models have demonstrated excellent performance in zero-shot and few-shot learning scenarios. In this study, we explore the potential of foundation models in plant disease recognition by proposing an efficient few-shot plant disease recognition model (PlantCaFo) based on foundation models. This model operates on an end-to-end network structure, integrating prior knowledge from multiple pretraining models. Specifically, we design a lightweight dilated contextual adapter (DCon-Adapter) to learn new knowledge from training data and use a weight decomposition matrix (WDM) to update the text weights. We test the proposed model on a public dataset, PlantVillage, and show that the model achieves an accuracy of 93.53 % in a “38-way 16-shot” setting. In addition, we conduct experiments on images collected from natural environments (Cassava dataset), achieving an accuracy improvement of 6.80 % over the baseline. To validate the model's generalization performance, we prepare an out-of-distribution dataset with 21 categories, and our model notably increases the accuracy of this dataset. Extensive experiments demonstrate that our model exhibits superior performance over other models in few-shot plant disease identification.