Soybean Leaf Disease Classification
This repository contains the source code, experimental configurations, algorithmic pseudocode, evaluation scripts, and reference results required to support the reproducibility of the research paper entitled "Hybrid Local–Global Feature Learning for Soyabean Leaf Disease Classification Using EfficientNetV2 and Swin Transformer". The proposed framework integrates EfficientNetV2 and Swin Transformer to learn complementary local and global features for soybean leaf disease classification. The repository includes Python implementation files, model architecture, dataset handling and preprocessing modules, training and evaluation scripts, cross-dataset validation procedures, ablation study implementations, statistical analysis scripts, and experimental configuration files. It also provides manuscript-reported reference results, including classification performance, baseline comparisons, cross-dataset generalization results, attention-fusion ablation studies, and computational efficiency measurements. The purpose of this repository is to support transparency, facilitate independent evaluation, and improve the reproducibility of the experimental methodology and findings reported in the manuscript. Associated manuscript: Hybrid Local–Global Feature Learning for Soyabean Leaf Disease Classification Using EfficientNetV2 and Swin Transformer.
Authors
- Parul Dubey
- Amol Chokhat
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2026-11-18
- DOI
- https://doi.org/10.5281/zenodo.24339850
- Primary Topic
- Smart Agriculture and AI
- Type
- article
- Field-Weighted Citation Impact
- 0.00