{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,30]],"date-time":"2026-09-30T06:57:03Z","timestamp":1790751423700,"version":"4.1.0"},"reference-count":52,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,9,30]],"date-time":"2026-09-30T00:00:00Z","timestamp":1790726400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Plant Sci."],"abstract":"<jats:p>Soybean growth habit is generally classified into determinate, indeterminate, and semi-determinate types, and its accurate recognition is important for breeding selection, high-density cultivation, and agronomic management. However, the semi-determinate type is intermediate between determinate and indeterminate types and shows continuous transitional characteristics with the indeterminate type in stem elongation, upper internode length, and pod distribution, making fine-grained discrimination between these two types difficult. Practical deployment also requires lightweight, efficient models. This study developed a lightweight fine-grained recognition framework for soybean growth habit from standardized plant images. The framework used RepViT-M1 as an efficient feature extraction backbone and integrated a multi-scale efficient channel attention module (MS-ECA) to enhance discriminative channel features, a Fine-grained Disambiguation Adapter (FDA) to recalibrate non-determinate features using class prototypes and reduce indeterminate\/semi-determinate confusion, and a hierarchical classification head to incorporate the agronomic hierarchy between determinate and non-determinate types. The dataset comprised 944 original soybean plant images and was expanded to 4,709 images through offline augmentation. The model was evaluated using five-fold cross-validation and an independent test set. The proposed model achieved 97.45% accuracy on the independent test set, improving the RepViT-M1 baseline by approximately 3 percentage points. Accuracy for the indeterminate type increased from 91.30% to 97.83%, indicating that the model effectively reduced confusion between indeterminate and semi-determinate types. The model had 4.94 M parameters, required 0.82 GFLOPs, and reached 108.75 FPS, demonstrating a favorable balance between accuracy and efficiency. The proposed framework improves fine-grained soybean growth habit recognition at low computational cost and demonstrates potential for standardized, high-throughput phenotypic identification through on-device Android inference.<\/jats:p>","DOI":"10.3389\/fpls.2026.1954156","type":"journal-article","created":{"date-parts":[[2026,9,30]],"date-time":"2026-09-30T06:06:01Z","timestamp":1790748361000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Soy-RepViT: a lightweight deep learning framework for fine-grained recognition of soybean growth habits"],"prefix":"10.3389","volume":"17","author":[{"given":"Dongbo","family":"Liu","sequence":"first","affiliation":[{"name":"College of Intelligence Science and Engineering, Northeast Agricultural University, Harbin","place":["Heilongjiang, China"]},{"name":"State Key Laboratory of Smart Farm Technologies and Systems","place":["Harbin, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Daohan","family":"Cui","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Smart Farm Technologies and Systems","place":["Harbin, China"]},{"name":"College of Agriculture, Northeast Agricultural University","place":["Harbin, Heilongjiang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hang","family":"Liu","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Smart Farm Technologies and Systems","place":["Harbin, China"]},{"name":"College of Agriculture, Northeast Agricultural University","place":["Harbin, Heilongjiang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tianyuan","family":"Song","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Smart Farm Technologies and Systems","place":["Harbin, China"]},{"name":"School of Resources and Environment, Northeast Agricultural University, Harbin","place":["Heilongjiang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhanguo","family":"Zhang","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Smart Farm Technologies and Systems","place":["Harbin, China"]},{"name":"College of Agriculture, Northeast Agricultural University","place":["Harbin, Heilongjiang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Qingshan","family":"Chen","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Smart Farm Technologies and Systems","place":["Harbin, China"]},{"name":"College of Agriculture, Northeast Agricultural University","place":["Harbin, Heilongjiang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Fanchuang","family":"Li","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Smart Farm Technologies and Systems","place":["Harbin, China"]},{"name":"College of Agriculture, Northeast Agricultural University","place":["Harbin, Heilongjiang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Rongsheng","family":"Zhu","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Smart Farm Technologies and Systems","place":["Harbin, China"]},{"name":"College of Agriculture, Northeast Agricultural University","place":["Harbin, Heilongjiang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,9,30]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"52","DOI":"10.1016\/j.tplants.2013.09.008","article-title":"Field high-throughput phenotyping: The new crop breeding frontier","volume":"19","author":"Araus","year":"2014","journal-title":"Trends Plant Sci."},{"key":"B2","doi-asserted-by":"publisher","first-page":"235","DOI":"10.2135\/cropsci1972.0011183x001200020028x","article-title":"Two genes affecting stem termination in soybeans","volume":"12","author":"Bernard","year":"1972","journal-title":"Crop Sci."},{"key":"B3","doi-asserted-by":"publisher","DOI":"10.3389\/fpls.2023.1173036","article-title":"A transformer-based approach for early prediction of soybean yield using time-series images","volume":"14","author":"Bi","year":"2023","journal-title":"Front. Plant Sci."},{"key":"B4","doi-asserted-by":"publisher","first-page":"11554","DOI":"10.1038\/s41598-022-15163-0","article-title":"A deep learning based approach for automated plant disease classification using vision transformer","volume":"12","author":"Borhani","year":"2022","journal-title":"Sci. Rep."},{"key":"B5","doi-asserted-by":"publisher","first-page":"137","DOI":"10.1007\/s10462-024-10759-6","article-title":"A survey on imbalanced learning: Latest research, applications and future directions","volume":"57","author":"Chen","year":"2024","journal-title":"Artif. Intell. Rev."},{"key":"B6","first-page":"5157","article-title":"Destruction and construction learning for fine-grained image recognition","author":"Chen","year":"2019"},{"key":"B7","doi-asserted-by":"publisher","DOI":"10.1016\/j.compag.2023.108072","article-title":"Plant image recognition with deep learning: A review","volume":"212","author":"Chen","year":"2023","journal-title":"Comput. Electron. Agric."},{"key":"B8","doi-asserted-by":"publisher","first-page":"11804","DOI":"10.1007\/s10489-024-05747-w","article-title":"A comprehensive review of model compression techniques in machine learning","volume":"54","author":"Dantas","year":"2024","journal-title":"Appl. Intell."},{"key":"B9","doi-asserted-by":"publisher","DOI":"10.3389\/fpls.2019.00508","article-title":"Leveraging image analysis for high-throughput plant phenotyping","volume":"10","author":"Das Choudhury","year":"2019","journal-title":"Front. Plant Sci."},{"key":"B10","first-page":"248","article-title":"ImageNet: A large-scale hierarchical image database","author":"Deng","year":"2009"},{"key":"B11","article-title":"An image is worth 16x16 words: Transformers for image recognition at scale","author":"Dosovitskiy","year":"2021"},{"key":"B12","doi-asserted-by":"publisher","first-page":"31","DOI":"10.1007\/s10618-010-0175-9","article-title":"A survey of hierarchical classification across different application domains","volume":"22","author":"Freitas","year":"2011","journal-title":"Data Min. Knowl. Discov."},{"key":"B13","first-page":"770","article-title":"Deep residual learning for image recognition","author":"He","year":"2016"},{"key":"B14","doi-asserted-by":"publisher","first-page":"1855","DOI":"10.2135\/cropsci2004.1855","article-title":"Effect of soybean stem growth habit on height and node number after beginning bloom in the midsouthern USA","volume":"44","author":"Heatherly","year":"2004","journal-title":"Crop Sci."},{"key":"B15","first-page":"13713","article-title":"Coordinate attention for efficient mobile network design","author":"Hou","year":"2021"},{"key":"B16","first-page":"1314","article-title":"Searching for mobilenetv3","author":"Howard","year":"2019"},{"key":"B17","first-page":"7132","article-title":"Squeeze-and-excitation networks","author":"Hu","year":"2018"},{"key":"B18","doi-asserted-by":"publisher","first-page":"285","DOI":"10.1007\/s10462-024-10922-z","article-title":"Domain generalization through meta-learning: A survey","volume":"57","author":"Khoee","year":"2024","journal-title":"Artif. Intell. Rev."},{"key":"B19","first-page":"1137","article-title":"A study of cross-validation and bootstrap for accuracy estimation and model selection","author":"Kohavi","year":"1995"},{"key":"B20","first-page":"1097","article-title":"ImageNet classification with deep convolutional neural networks","volume-title":"Advances in Neural Information Processing Systems","author":"Krizhevsky","year":"2012"},{"key":"B21","doi-asserted-by":"publisher","DOI":"10.3389\/fpls.2024.1477616","article-title":"Molecular and genetic basis of plant architecture in soybean","volume":"15","author":"Li","year":"2024","journal-title":"Front. Plant Sci."},{"key":"B22","doi-asserted-by":"publisher","DOI":"10.1016\/j.compag.2023.108166","article-title":"Flooding-based MobileNet to identify cucumber diseases from leaf images in natural scenes","volume":"213","author":"Liu","year":"2023","journal-title":"Comput. Electron. Agric."},{"key":"B23","doi-asserted-by":"publisher","first-page":"198","DOI":"10.1104\/pp.109.150607","article-title":"The soybean stem growth habit gene Dt1 is an ortholog of Arabidopsis TERMINAL FLOWER1","volume":"153","author":"Liu","year":"2010","journal-title":"Plant Physiol."},{"key":"B24","doi-asserted-by":"publisher","first-page":"3375","DOI":"10.3390\/agriengineering6030193","article-title":"Image analysis artificial intelligence technologies for plant phenotyping: Current state of the art","volume":"6","author":"Maraveas","year":"2024","journal-title":"AgriEngineering"},{"key":"B25","doi-asserted-by":"publisher","DOI":"10.1016\/j.isci.2023.108709","article-title":"Computer vision-based plants phenotyping: A comprehensive survey","volume":"27","author":"Meraj","year":"2024","journal-title":"iScience"},{"key":"B26","doi-asserted-by":"publisher","DOI":"10.3389\/fbinf.2024.1457619","article-title":"A review of model evaluation metrics for machine learning in genetics and genomics","volume":"4","author":"Miller","year":"2024","journal-title":"Front. Bioinf."},{"key":"B27","doi-asserted-by":"publisher","DOI":"10.3389\/fpls.2016.01419","article-title":"Using deep learning for image-based plant disease detection","volume":"7","author":"Mohanty","year":"2016","journal-title":"Front. Plant Sci."},{"key":"B28","doi-asserted-by":"publisher","first-page":"771","DOI":"10.1146\/annurev-arplant-070523-042828","article-title":"Deep learning in image-based plant phenotyping","volume":"75","author":"Murphy","year":"2024","journal-title":"Annu. Rev. Plant Biol."},{"key":"B29","doi-asserted-by":"publisher","first-page":"820","DOI":"10.1162\/tacl_a_00675","article-title":"A closer look at classification evaluation metrics and a critical reflection of common evaluation practice","volume":"12","author":"Opitz","year":"2024","journal-title":"Trans. Assoc. For. Comput. Linguistics"},{"key":"B30","doi-asserted-by":"publisher","DOI":"10.3390\/rs15112891","article-title":"Deep-learning-based rice phenological stage recognition","volume":"15","author":"Qin","year":"2023","journal-title":"Remote Sens."},{"key":"B31","doi-asserted-by":"publisher","first-page":"6086","DOI":"10.1038\/s41598-024-56706-x","article-title":"Evaluation metrics and statistical tests for machine learning","volume":"14","author":"Rainio","year":"2024","journal-title":"Sci. Rep."},{"key":"B32","doi-asserted-by":"publisher","DOI":"10.1016\/j.compag.2024.109108","article-title":"High-throughput proximal ground crop phenotyping systems: A comprehensive review","volume":"224","author":"Rui","year":"2024","journal-title":"Comput. Electron. Agric."},{"key":"B33","first-page":"618","article-title":"Grad-CAM: Visual explanations from deep networks via gradient-based localization","author":"Selvaraju","year":"2017"},{"key":"B34","first-page":"1","article-title":"Very deep convolutional networks for large-scale image recognition","author":"Simonyan","year":"2015"},{"key":"B35","first-page":"4077","article-title":"Prototypical networks for few-shot learning","volume-title":"Advances in Neural Information Processing Systems","author":"Snell","year":"2017"},{"key":"B36","doi-asserted-by":"publisher","DOI":"10.3390\/agronomy15081831","article-title":"Empowering smart soybean farming with deep learning: Progress, challenges, and future perspectives","volume":"15","author":"Sun","year":"2025","journal-title":"Agronomy"},{"key":"B37","first-page":"6105","article-title":"EfficientNet: Rethinking model scaling for convolutional neural networks","author":"Tan","year":"2019"},{"key":"B38","doi-asserted-by":"publisher","DOI":"10.3389\/fpls.2017.01190","article-title":"Deep plant phenomics: A deep learning platform for complex plant phenotyping tasks","volume":"8","author":"Ubbens","year":"2017","journal-title":"Front. Plant Sci."},{"key":"B39","first-page":"5785","article-title":"FastViT: A fast hybrid vision transformer using structural reparameterization","author":"Vasu","year":"2023"},{"key":"B40","first-page":"15909","article-title":"RepViT: Revisiting mobile CNN from ViT perspective","author":"Wang","year":"2024"},{"key":"B41","doi-asserted-by":"publisher","DOI":"10.3389\/fpls.2025.1614898","article-title":"A study on the non-contact measurement of sunflower disk inclination and its application to accurate phenotypic analysis","volume":"16","author":"Wang","year":"2025","journal-title":"Front. Plant Sci."},{"key":"B42","doi-asserted-by":"publisher","DOI":"10.3389\/fpls.2022.730190","article-title":"Corn seed defect detection based on watershed algorithm and two-pathway convolutional neural networks","volume":"13","author":"Wang","year":"2022","journal-title":"Front. Plant Sci."},{"key":"B43","first-page":"11534","article-title":"ECA-Net: Efficient channel attention for deep convolutional neural networks","author":"Wang","year":"2020"},{"key":"B44","doi-asserted-by":"publisher","first-page":"1642","DOI":"10.1016\/j.cj.2023.08.002","article-title":"Regulation of soybean stem growth habit: A ten-year progress report","volume":"11","author":"Xiong","year":"2023","journal-title":"Crop J."},{"key":"B45","first-page":"2740","article-title":"HD-CNN: Hierarchical deep convolutional neural networks for large scale visual recognition","author":"Yan","year":"2015"},{"key":"B46","doi-asserted-by":"publisher","first-page":"2805","DOI":"10.1007\/s10115-023-01853-2","article-title":"A survey of automated data augmentation algorithms for deep learning-based image classification tasks","volume":"65","author":"Yang","year":"2023","journal-title":"Knowl. Inf. Syst."},{"key":"B47","doi-asserted-by":"publisher","DOI":"10.3897\/bdj.8.e57090","article-title":"Detection and annotation of plant organs from digitised herbarium scans using deep learning","volume":"8","author":"Younis","year":"2020","journal-title":"Biodivers. Data J."},{"key":"B48","doi-asserted-by":"publisher","first-page":"6190","DOI":"10.3934\/mbe.2024272","article-title":"Image data augmentation techniques based on deep learning: A survey","volume":"21","author":"Zeng","year":"2024","journal-title":"Math. Biosci. Eng."},{"key":"B49","doi-asserted-by":"publisher","DOI":"10.3390\/plants13182613","article-title":"High-precision automated soybean phenotypic feature extraction based on deep learning and computer vision","volume":"13","author":"Zhang","year":"2024","journal-title":"Plants"},{"key":"B50","first-page":"6848","article-title":"ShuffleNet: An extremely efficient convolutional neural network for mobile devices","author":"Zhang","year":"2018"},{"key":"B51","doi-asserted-by":"publisher","DOI":"10.1016\/j.compag.2021.106230","article-title":"Real-time recognition system of soybean seed full-surface defects based on deep learning","volume":"187","author":"Zhao","year":"2021","journal-title":"Comput. Electron. Agric."},{"key":"B52","doi-asserted-by":"publisher","first-page":"1569","DOI":"10.1016\/j.cj.2023.04.012","article-title":"SPP-extractor: Automatic phenotype extraction for densely grown soybean plants","volume":"11","author":"Zhou","year":"2023","journal-title":"Crop J."}],"container-title":["Frontiers in Plant Science"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpls.2026.1954156\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,30]],"date-time":"2026-09-30T06:06:02Z","timestamp":1790748362000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpls.2026.1954156\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,30]]},"references-count":52,"alternative-id":["10.3389\/fpls.2026.1954156"],"URL":"https:\/\/doi.org\/10.3389\/fpls.2026.1954156","relation":{},"ISSN":["1664-462X"],"issn-type":[{"value":"1664-462X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,30]]},"article-number":"1954156"}}