Lightweight Cross-Domain Few-Shot Plant Disease Recognition Through Target-Domain Statistical Calibration.

Abstract

Plant disease recognition models trained under laboratory conditions often degrade markedly after cross-domain transfer because of the pronounced distribution gap between source and target domains and the scarcity of labeled target-domain samples. To address the transfer task from PlantVillage (PV_100) to PlantDoc, this study develops and evaluates a lightweight cross-domain few-shot plant disease recognition method under a strict PlantVillage-to-PlantDoc protocol. The method integrates EfficientNet-B0 feature extraction, cosine-similarity-based prototypical classification, and training-time target-domain BN adaptation (TBA). During training, unlabeled target-domain images are used only for BN statistical calibration, whereas inference is limited to feature extraction and prototype matching, without gradient updates or iterative optimization. Under a unified experimental protocol, the proposed method achieved cross-split mean accuracies of 42.69 ± 0.62% for one-shot and 54.24 ± 0.72% for five-shot, where ± denotes the standard deviation across three strict data splits; it outperformed ProtoNet by 7.44 and 9.43 percentage points, respectively. Ablation results indicate that TBA is the main source of performance improvement, whereas more complex adaptation strategies do not yield stable additional gains. The core encoder can be executed entirely on the NPU, with an estimated single-sample inference latency as low as 0.658 ms, indicating strong potential for encoder-level mobile deployment.

Plant phenotyping relevance

植物病害状態を画像から認識する手法を開発し、厳密なクロスドメイン条件、比較、アブレーションで評価しており、病害表現型の取得・推定が研究の中心である。

Code and data availability

The article uses public PlantVillage and PlantDoc datasets, but these are cited prior-work datasets rather than paper-specific deposits. No author code, trained model checkpoints, data splits, or supplement with phenotyping assets is described; no code or data availability statement with a public URL appears in the-sup

No evidence-backed public reproduction asset is currently recorded.