{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,21]],"date-time":"2026-08-21T06:24:17Z","timestamp":1787293457594,"version":"build-2736575974"},"reference-count":55,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,8,21]],"date-time":"2026-08-21T00:00:00Z","timestamp":1787270400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100012166","name":"National Key Research and Development Program of China","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100012166","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Plant Sci."],"abstract":"<jats:p>Accurate maize growth stage recognition is critical for precision agricultural management, but it remains challenging due to subtle morphological differences between adjacent stages, small seedling targets, and substantial cross-year environmental variability. To address these challenges, this study proposes a Multi-Scale Temporal Feature Fusion Network, termed MS- TFNet, for image-level maize growth stage recognition using red-green-blue (RGB) images and acquisition-date information. The proposed network integrates three complementary components. First, a Multi-Scale Feature Extraction Module (MS) is introduced to capture maize structural variations at different receptive-field scales. Second, a Vegetation-Aware Convolutional Block Attention Module (VCBAM) is designed to adaptively refine RGB feature responses and improve the representation of maize-related visual cues under complex field backgrounds. Third, a Temporal Feature Module (TFM) encodes image acquisition dates using date-based sine\u2013cosine temporal encoding, enabling phenological prior information to be incorporated into single-image growth stage classification. Experiments were conducted on a long-term field observation dataset containing 24,239 maize images collected over eight growing seasons and covering nine maize growth stages. The results show that MS-TFNet achieved a mean Overall Accuracy (OA) of 83.64%, outperforming the ResNet-50 baseline by 4.46 percentage points. Under a \u00b13-day temporal tolerance criterion, the recognition accuracy further increased to 93.76%, indicating improved robustness for practical crop monitoring scenarios. Overall, MS-TFNet provides an effective framework for maize growth stage recognition under complex field conditions by jointly exploiting multi-scale visual representation, vegetation-aware attention, and acquisition-date-based phenological priors.<\/jats:p>","DOI":"10.3389\/fpls.2026.1882260","type":"journal-article","created":{"date-parts":[[2026,8,21]],"date-time":"2026-08-21T05:37:03Z","timestamp":1787290623000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["MS-TFNet: a multi-scale temporal feature fusion network for maize growth stage recognition"],"prefix":"10.3389","volume":"17","author":[{"given":"Su","family":"Wu","sequence":"first","affiliation":[{"name":"College of Computer Science and Technology, Harbin Engineering University","place":["Harbin, China"]},{"name":"Henan Zhongyuan Photoelectric Measurement and Control Technology Co., Ltd.","place":["Zhengzhou, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhihao","family":"Liu","sequence":"additional","affiliation":[{"name":"Henan Zhongyuan Photoelectric Measurement and Control Technology Co., Ltd.","place":["Zhengzhou, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dongli","family":"Wu","sequence":"additional","affiliation":[{"name":"Meteorological Observation Centre, China Meteorological Administration","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yijun","family":"Liu","sequence":"additional","affiliation":[{"name":"Henan Zhongyuan Photoelectric Measurement and Control Technology Co., Ltd.","place":["Zhengzhou, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hanxiao","family":"Zhang","sequence":"additional","affiliation":[{"name":"Henan Zhongyuan Photoelectric Measurement and Control Technology Co., Ltd.","place":["Zhengzhou, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mengxuan","family":"Liu","sequence":"additional","affiliation":[{"name":"Henan Zhongyuan Photoelectric Measurement and Control Technology Co., Ltd.","place":["Zhengzhou, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dongzhe","family":"Tian","sequence":"additional","affiliation":[{"name":"Henan Zhongyuan Photoelectric Measurement and Control Technology Co., Ltd.","place":["Zhengzhou, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Guisheng","family":"Yin","sequence":"additional","affiliation":[{"name":"College of Computer Science and Technology, Harbin Engineering University","place":["Harbin, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,8,21]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","DOI":"10.3390\/agriculture15141523","article-title":"Agrifusionnet: a lightweight deep learning model for multisource plant disease diagnosis","volume":"15","author":"Albahli","year":"2025","journal-title":"Agriculture"},{"key":"B2","doi-asserted-by":"publisher","DOI":"10.1016\/j.ecoinf.2023.102217","article-title":"A review of deep learning techniques used in agriculture","volume":"77","author":"Attri","year":"2023","journal-title":"Ecol. Inf."},{"key":"B3","doi-asserted-by":"publisher","first-page":"171","DOI":"10.1016\/0308-521X(96)00011-X","article-title":"The \u2018school of de wit\u2019 crop growth simulation models: a pedigree and historical overview","volume":"52","author":"Bouman","year":"1996","journal-title":"Agric. Syst."},{"key":"B4","doi-asserted-by":"publisher","DOI":"10.1016\/j.compag.2023.107881","article-title":"Attention-aided semantic segmentation network for weed identification in pineapple field","volume":"210","author":"Cai","year":"2023","journal-title":"Comput. Electron. Agric."},{"key":"B5","first-page":"1597","article-title":"A simple framework for contrastive learning of visual representations","volume-title":"Pmlr","author":"Chen","year":"2020"},{"key":"B6","doi-asserted-by":"publisher","first-page":"57","DOI":"10.32604\/cmc.2025.066056","article-title":"A systematic review of deep learning-based object detection in agriculture: methods, challenges, and future directions","volume":"84","author":"Dalal","year":"2025","journal-title":"Computers, Materials & Continua"},{"key":"B7","doi-asserted-by":"publisher","first-page":"187","DOI":"10.1007\/s11119-022-09938-8","article-title":"Uav-based multi-sensor data fusion and machine learning algorithm for yield prediction in wheat","volume":"24","author":"Fei","year":"2023","journal-title":"Precis. Agric."},{"key":"B8","doi-asserted-by":"publisher","DOI":"10.3389\/fpls.2025.1639533","article-title":"Field-deployable lightweight yolov8n for real-time detection and counting of maize seedlings using uav rgb imagery","volume":"16","author":"Feng","year":"2025","journal-title":"Front. Plant Sci."},{"key":"B9","doi-asserted-by":"publisher","first-page":"961","DOI":"10.1016\/j.cj.2025.03.012","article-title":"Dynamic UAV data fusion and deep learning for improved maize phenological-stage tracking","volume":"13","author":"Feng","year":"2025","journal-title":"Crop J."},{"key":"B10","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1406.2661","article-title":"Generative adversarial nets","author":"Goodfellow","year":"2014","journal-title":"arXiv preprint arXiv:1406.2661"},{"key":"B11","first-page":"346","article-title":"Spatial pyramid pooling in deep convolutional networks for visual recognition","volume-title":"Springer","author":"He","year":"2014"},{"key":"B12","first-page":"770","article-title":"Deep residual learning for image recognition","author":"He","year":"2016"},{"key":"B13","first-page":"1","article-title":"mixup: beyond empirical risk minimization","volume-title":"Proceedings of the 6th International Conference on Learning Representations (ICLR)","author":"Hongyi","year":"2018"},{"key":"B14","first-page":"7132","article-title":"Squeeze-and-excitation networks","volume-title":"Ieee","author":"Hu","year":"2018"},{"key":"B15","doi-asserted-by":"publisher","DOI":"10.1016\/j.atech.2025.101202","article-title":"CornMFN: a multimodal fusion network for corn phenology stage identification","volume":"12","author":"Huang","year":"2025","journal-title":"Smart Agric. Technol."},{"key":"B16","doi-asserted-by":"publisher","first-page":"833","DOI":"10.1016\/j.cageo.2004.05.006","article-title":"Timesat\u2013a program for analyzing time-series of satellite sensor data","volume":"30","author":"J\u00f6nsson","year":"2004","journal-title":"Comput. Geosciences"},{"key":"B17","doi-asserted-by":"publisher","DOI":"10.1016\/j.compag.2024.108690","article-title":"Harnessing synthetic data for enhanced detection of pine wilt disease: an image classification approach","volume":"218","author":"Jung","year":"2024","journal-title":"Comput. Electron. Agric."},{"key":"B18","doi-asserted-by":"publisher","first-page":"70","DOI":"10.1016\/j.compag.2018.02.016","article-title":"Deep learning in agriculture: a survey","volume":"147","author":"Kamilaris","year":"2018","journal-title":"Comput. Electron. Agric."},{"key":"B19","doi-asserted-by":"publisher","DOI":"10.3390\/agriculture15131351","article-title":"Objectdetection in agriculture: a comprehensive review of methods, applications, challenges, and future directions","volume":"15","author":"Khan","year":"2025","journal-title":"Agriculture"},{"key":"B20","doi-asserted-by":"publisher","DOI":"10.3389\/fpls.2025.1707985","article-title":"Instance-level phenotype-based growth stage classification of basil in multi-plant environments","volume":"16","author":"Kim","year":"2025","journal-title":"Front. Plant Sci."},{"key":"B21","doi-asserted-by":"publisher","first-page":"245","DOI":"10.1016\/j.isprsjprs.2019.02.013","article-title":"Deep learning for real-time fruit detection and counting in orchards","volume":"150","author":"Koirala","year":"2019","journal-title":"ISPRS J. Photogramm. Remote Sens."},{"key":"B23","doi-asserted-by":"publisher","first-page":"100144","DOI":"10.1016\/j.plaphe.2025.100144","article-title":"Wpdsi: a deep learning method for wheat phenology detection from single-temporal images","volume":"7","author":"Li","year":"2025","journal-title":"Plant Phenomics"},{"key":"B24","doi-asserted-by":"publisher","DOI":"10.1016\/j.eja.2024.127423","article-title":"Wheat growth stage identification method based on multimodal data","volume":"162","author":"Li","year":"2025","journal-title":"Eur. J. Agron."},{"key":"B22","doi-asserted-by":"publisher","first-page":"1676","DOI":"10.1016\/j.agwat.2010.05.025","article-title":"Root growth, available soil water, and water-use efficiency of winter wheat under different irrigation regimes applied at different growth stages in north China","volume":"97","author":"Li","year":"2010","journal-title":"Agric. Water Manage."},{"key":"B25","first-page":"2117","article-title":"Feature pyramid networks for object detection","volume-title":"Ieee","author":"Lin","year":"2017"},{"key":"B26","first-page":"2980","article-title":"Focal loss for dense object detection","author":"Lin","year":"2017"},{"key":"B27","doi-asserted-by":"publisher","DOI":"10.1016\/j.compag.2024.108668","article-title":"Gldcnet: a novel convolutional neural network for grapevine leafroll disease recognition using uav-based imagery","volume":"218","author":"Liu","year":"2024","journal-title":"Comput. Electron. Agric."},{"key":"B28","doi-asserted-by":"publisher","DOI":"10.3389\/fpls.2017.02106","article-title":"Maize yield and planting date relationship: a synthesis-analysis for us high-yielding contest-winner and field research data","volume":"8","author":"Long","year":"2017","journal-title":"Front. Plant Sci."},{"key":"B29","doi-asserted-by":"publisher","DOI":"10.1016\/j.compag.2024.108663","article-title":"Enhancing fusarium head blight detection in wheat crops using hyperspectral indices and machine learning classifiers","volume":"218","author":"Mustafa","year":"2024","journal-title":"Comput. Electron. Agric."},{"key":"B30","doi-asserted-by":"publisher","DOI":"10.3390\/rs16142672","article-title":"A cnn-and self-attention-based maize growth stage recognition method and platform from uav orthophoto images","volume":"16","author":"Ni","year":"2024","journal-title":"Remote Sens."},{"key":"B31","doi-asserted-by":"publisher","first-page":"119","DOI":"10.7744\/kjoas.20200004","article-title":"Digital image-based plant phenotyping: a review","volume":"47","author":"Omari","year":"2020","journal-title":"Korean J. Agric. Sci."},{"key":"B32","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1016\/j.biosystemseng.2019.05.002","article-title":"Maize seedling detection under different growth stages and complex field environments based on an improved faster r\u2013cnn","volume":"184","author":"Quan","year":"2019","journal-title":"Biosyst. Eng."},{"key":"B33","doi-asserted-by":"publisher","DOI":"10.3390\/ijgi7040129","article-title":"Multi-temporal land cover classification with sequential recurrent encoders","volume":"7","author":"Ru\u00dfwurm","year":"2018","journal-title":"ISPRS Int. J. Geo-Inf."},{"key":"B34","doi-asserted-by":"publisher","DOI":"10.1016\/j.compag.2023.107944","article-title":"Comparison of deep learning methods for grapevine growth stage recognition","volume":"211","author":"Schieck","year":"2023","journal-title":"Comput. Electron. Agric."},{"key":"B35","doi-asserted-by":"publisher","DOI":"10.1016\/j.compag.2022.107217","article-title":"Technological revolutions in smart farming: current trends, challenges & future directions","volume":"201","author":"Sharma","year":"2022","journal-title":"Comput. Electron. Agric."},{"key":"B36","doi-asserted-by":"publisher","DOI":"10.1080\/08839514.2025.2469372","article-title":"Msnet: a novel deep learning framework for efficient missing seedling detection in maize fields","volume":"39","author":"Shi","year":"2025","journal-title":"Appl. Artif. Intell."},{"key":"B37","doi-asserted-by":"publisher","first-page":"470","DOI":"10.1016\/j.asr.2019.09.034","article-title":"Monitoring of maize lodging using multi-temporal sentinel-1 sar data","volume":"65","author":"Shu","year":"2020","journal-title":"Adv. Space Res."},{"key":"B38","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1409.1556","article-title":"Very deep convolutional networks for large-scale image recognition","author":"Simonyan","year":"2014","journal-title":"arXiv preprint arXiv:1409.1556"},{"key":"B39","doi-asserted-by":"publisher","first-page":"53","DOI":"10.1016\/j.tplants.2020.07.010","article-title":"Challenges and opportunities in machine-augmented plant stress phenotyping","volume":"26","author":"Singh","year":"2021","journal-title":"Trends Plant Sci."},{"key":"B40","doi-asserted-by":"publisher","DOI":"10.3389\/fpls.2026.1730706","article-title":"Adaptive multi-scale feature refinement for wheat phenology recognition using cross-scale attention mechanisms","volume":"17","author":"Sun","year":"2026","journal-title":"Front. Plant Sci."},{"key":"B41","doi-asserted-by":"publisher","first-page":"1","DOI":"10.6041\/j.issn.1000-1298.2020.05.001","article-title":"Research progress of image sensing and deep learning in agriculture","volume":"51","author":"Sun","year":"2020","journal-title":"Trans. Chin. Soc. For. Agric. Machinery"},{"key":"B42","first-page":"1","article-title":"Going deeper with convolutions","author":"Szegedy","year":"2015"},{"key":"B43","first-page":"6105","article-title":"Efficientnet: rethinking model scaling for convolutional neural networks","volume-title":"Pmlr","author":"Tan","year":"2019"},{"key":"B44","doi-asserted-by":"publisher","first-page":"117","DOI":"10.1016\/j.isprsjprs.2024.01.025","article-title":"Deep learning with multi-scale temporal hybrid structure for robust crop mapping","volume":"209","author":"Tang","year":"2024","journal-title":"ISPRS J. Photogramm. Remote Sens."},{"key":"B45","doi-asserted-by":"publisher","first-page":"201","DOI":"10.1016\/S1161-0301(02)00106-5","article-title":"On approaches and applications of the wageningen crop models","volume":"18","author":"van Ittersum","year":"2003","journal-title":"Eur. J. Agron."},{"key":"B46","doi-asserted-by":"publisher","DOI":"10.3390\/agronomy15010013","article-title":"Determination of wheat growth stages using image sequences and deep learning","volume":"15","author":"Wang","year":"2025","journal-title":"Agronomy"},{"key":"B47","doi-asserted-by":"publisher","first-page":"116663","DOI":"10.1109\/ACCESS.2020.3005150","article-title":"Multi-scale cnn for fine-grained image recognition","volume":"8","author":"Won","year":"2020","journal-title":"IEEE Access"},{"key":"B48","first-page":"3","article-title":"Cbam: convolutional block attention module","volume-title":"Springer","author":"Woo","year":"2018"},{"key":"B49","doi-asserted-by":"publisher","DOI":"10.3390\/electronics10010081","article-title":"A review of plant phenotypic image recognition technology based on deep learning","volume":"10","author":"Xiong","year":"2021","journal-title":"Electronics"},{"key":"B50","doi-asserted-by":"publisher","first-page":"43","DOI":"10.7671\/j.issn.1001-411X.202209042","article-title":"Research status and prospect of crop information monitoring technology in field","volume":"44","author":"Xuejun","year":"2023","journal-title":"J. South. China Agric. Univ."},{"key":"B51","doi-asserted-by":"publisher","DOI":"10.1016\/j.compag.2024.109728","article-title":"Phenologynet: A fine-grained approach for crop-phenology classification fusing convolutional neural network and phenotypic similarity","volume":"229","author":"Yang","year":"2025","journal-title":"Comput. Electron. Agric."},{"key":"B52","article-title":"Multi-scale context aggregation by dilated convolutions","author":"Yu","year":"2016"},{"key":"B53","first-page":"6023","article-title":"Cutmix: Regularization strategy to train strong classifiers with localizable features","author":"Yun","year":"2019"},{"key":"B54","doi-asserted-by":"publisher","DOI":"10.1016\/j.jag.2024.104204","article-title":"A new attention-based deep metric model for crop type mapping in complex agricultural landscapes using multisource remote sensing data","volume":"134","author":"Zheng","year":"2024","journal-title":"Int. J. Appl. Earth Obs. Geoinf."},{"key":"B55","doi-asserted-by":"publisher","DOI":"10.1016\/j.agsy.2024.104214","article-title":"Yield prediction through uav-based multispectral imaging and deep learning in rice breeding trials","volume":"223","author":"Zhou","year":"2025","journal-title":"Agric. Syst."}],"container-title":["Frontiers in Plant Science"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpls.2026.1882260\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,21]],"date-time":"2026-08-21T05:37:08Z","timestamp":1787290628000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpls.2026.1882260\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,21]]},"references-count":55,"alternative-id":["10.3389\/fpls.2026.1882260"],"URL":"https:\/\/doi.org\/10.3389\/fpls.2026.1882260","relation":{},"ISSN":["1664-462X"],"issn-type":[{"value":"1664-462X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,21]]},"article-number":"1882260"}}