BLAP: lesion-aware adaptive multi-scale visual prompt tuning for few-shot crop disease diagnosis.

Abstract

Introduction Applying general-purpose vision-language models (VLMs) to crop disease diagnosis presents three critical bottlenecks: reliance on large-scale annotated data, the high computational cost of full finetuning, and existing adaptation methods designed mainly for discriminative classification without sufficient visual-linguistic interaction for generative diagnosis. Methods We propose BLAP, an adaptive multi-scale visual prompt fine-tuning framework built upon BLIP-2. BLAP introduces an adaptive visual prompt fusion module (APFM) with learnable prompt vectors and a gating mechanism, together with a multi-scale pyramid feature fusion module (PFM). All BLIP-2 backbone parameters are frozen, and only 0.11% of the model parameters are optimized. Results On a few-shot dataset comprising 990 images from 11 crops and 33 disease categories, BLAP achieved 92.78% recognition accuracy, outperforming the BLIP-2+LoRA baseline by 21.67 percentage points. BLEU-4 and ROUGE-L scores reached 0.6507 and 0.7184, respectively, while inference latency increased by only 2.15%. Discussion BLAP provides a lightweight solution that balances accuracy, efficiency, and interpretability for crop disease diagnosis in resource-constrained settings. The proposed dynamic prompt fusion and multiscale pyramid adaptation strategy may also be extended to parameter-efficient fine-tuning of visionlanguage models in other domain-specific applications.

Plant phenotyping relevance

作物病害画像から病徴・病害状態を推定する視覚モデル適応手法BLAPの開発と評価が中心であり、植物病害フェノタイピング手法に該当する。

Code and data availability

The paper's few-shot crop disease dataset (990 images, 11 crops, 33 classes) is compiled entirely from four public Mendeley Data image repositories, each cited in Table 1 as the data source for specific crop/disease classes. These are the plant image inputs directly used for this paper's phenotyping/analysis. No author