Lightweight frequency-aware hybrid vision transformer for fine-grained plant disease recognition
- 1School of Life Science, Guizhou Normal University
- 2School of Information Management and Artificial Intelligence, Zhejiang University of Finance and Economics
- 3School of Computer Science and Technology, Zhejiang University of Water Resources and Electric Power
Abstract
Plant diseases can substantially reduce crop yield and quality, while accurate disease recognition on edge devices remains challenging because images often contain subtle lesion textures and complex backgrounds. This study proposes a lightweight frequency-aware Dual-Embedding Hybrid Vision Transformer, termed DE-HVT, for fine-grained plant disease recognition. Here, “frequency-aware” does not denote an explicit or orthogonal separation into independent frequency bands. Instead, it refers to complementary frequency-biased processing by two spatially aligned embedding branches: OBPE preserves slowly varying structural cues, whereas the CDC-based TexPE branch has a zero response at zero spatial frequency under the linearised θ = 1 analysis and therefore exhibits difference-sensitive, high-pass-like responses to lesion boundaries and textures. Their outputs are fused by element-wise addition before MHSA, while ConvFFN introduces local spatial filtering after global token interaction. Experiments were conducted on the Plant Pathology and PlantVillage datasets using the same training and evaluation protocols for all compared models. Across three independent runs, DE-HVT achieved a mean Top-1 accuracy of 93.51 ± 0.44 % and a mean macro-averaged F1 score of 93.64 ± 0.40 % on the more challenging Plant Pathology dataset. On the controlled-background PlantVillage dataset, all evaluated models achieved near-ceiling recognition performance. The model contains 4.10 million parameters and requires 0.85 GFLOPs. On a Snapdragon 865-based mobile platform, it used 46.86 MB of memory and recorded a maximum observed inference latency of 111.96 ms per image. These results indicate that DE-HVT provides a favourable balance between recognition performance and computational efficiency within the evaluated datasets and under the tested mobile-device conditions.
Related
- Soy-RepViT: a lightweight deep learning framework for fine-grained recognition of soybean growth habits
- Lightweight CNN and transformer detectors for real-time single-fruit date-variety recognition on edge GPU hardware
- A lightweight privacy-aware federated vision transformer for tomato leaf disease classification
- MambaSeed: A lightweight Mamba-based hybrid vision network for tomato leaf disease recognition
- Fine-grained hierarchical crop type classification from integrated hyperspectral EnMAP data and multispectral sentinel-2 time series: A large-scale dataset and dual-stream transformer method
BibTeX
@article{li2026lightweight,
title = {{Lightweight frequency-aware hybrid vision transformer for fine-grained plant disease recognition}},
author = {Kun Li and Li Dong and Jing Qian and Shuai Wang and Luo-Luo Jiang and Wen-Jing Li},
journal = {Frontiers in Plant Science},
year = {2026},
month = oct,
doi = {10.3389/fpls.2026.1956762},
url = {https://www.frontiersin.org/articles/10.3389/fpls.2026.1956762/full},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Lightweight frequency-aware hybrid vision transformer for fine-grained plant disease recognition
AU - Kun Li
AU - Li Dong
AU - Jing Qian
AU - Shuai Wang
AU - Luo-Luo Jiang
AU - Wen-Jing Li
PY - 2026
DA - 2026/10/13
JO - Frontiers in Plant Science
DO - 10.3389/fpls.2026.1956762
UR - https://www.frontiersin.org/articles/10.3389/fpls.2026.1956762/full
AB - Plant diseases can substantially reduce crop yield and quality, while accurate disease recognition on edge devices remains challenging because images often contain subtle lesion textures and complex backgrounds. This study proposes a lightweight frequency-aware Dual-Embedding Hybrid Vision Transformer, termed DE-HVT, for fine-grained plant disease recognition. Here, “frequency-aware” does not denote an explicit or orthogonal separation into independent frequency bands. Instead, it refers to complementary frequency-biased processing by two spatially aligned embedding branches: OBPE preserves slowly varying structural cues, whereas the CDC-based TexPE branch has a zero response at zero spatial frequency under the linearised θ = 1 analysis and therefore exhibits difference-sensitive, high-pass-like responses to lesion boundaries and textures. Their outputs are fused by element-wise addition before MHSA, while ConvFFN introduces local spatial filtering after global token interaction. Experiments were conducted on the Plant Pathology and PlantVillage datasets using the same training and evaluation protocols for all compared models. Across three independent runs, DE-HVT achieved a mean Top-1 accuracy of 93.51 ± 0.44 % and a mean macro-averaged F1 score of 93.64 ± 0.40 % on the more challenging Plant Pathology dataset. On the controlled-background PlantVillage dataset, all evaluated models achieved near-ceiling recognition performance. The model contains 4.10 million parameters and requires 0.85 GFLOPs. On a Snapdragon 865-based mobile platform, it used 46.86 MB of memory and recorded a maximum observed inference latency of 111.96 ms per image. These results indicate that DE-HVT provides a favourable balance between recognition performance and computational efficiency within the evaluated datasets and under the tested mobile-device conditions.
SN - 1664-462X
ER -