{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,20]],"date-time":"2026-08-20T06:23:51Z","timestamp":1787207031666,"version":"build-2736575974"},"reference-count":48,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,8,20]],"date-time":"2026-08-20T00:00:00Z","timestamp":1787184000000},"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>Accurate tea cultivar identification is a long-standing challenge in tea plant propagation and cultivation. Traditional cultivar identification based on morphological observation relies heavily on expert experience, leading to strong subjectivity; while DNA fingerprinting-based approaches are time-consuming and costly, restricting their large-scale application. This study attempts to construct a deep learning- based on leaf morphological and visual characteristics. We collected an original dataset comprising 11,404 leaf images from 10 clonal tea cultivars for model development and evaluation. Comparative experiments were conducted using five models: ResNet50, EfficientNet-B3, MobileNetV3-large, ViT-B16, and ConvNeXt-Tiny. ResNet50 achieved the highest classification accuracy, whereas MobileNetV3-Large provided a better balance between accuracy and computational efficiency and was therefore selected as the baseline model for lightweight optimization. Channel pruning and PTDQ were further applied to compress and optimize MobileNetV3-Large, resulting in a lightweight and efficient tea cultivar identification model. The optimized model achieved a test accuracy of 97.15% on the independent testing set, while reducing the model size from 34.40 MB to 7.67 MB and decreasing inference latency from 42.35 ms to 27.67 ms. Future work will focus on integration into Android-based field applications. This study provides an efficient lightweight solution for practical tea cultivar identification and mobile deployment.<\/jats:p>","DOI":"10.3389\/fpls.2026.1863075","type":"journal-article","created":{"date-parts":[[2026,8,20]],"date-time":"2026-08-20T05:28:30Z","timestamp":1787203710000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["A lightweight deep learning framework for tea cultivar identification based on leaf images"],"prefix":"10.3389","volume":"17","author":[{"given":"Menghan","family":"He","sequence":"first","affiliation":[{"name":"Tea Resources Utilization and Quality Testing Key Laboratory of Sichuan Province, Sichuan Agricultural University","place":["Chengdu, China"]},{"name":"College of Horticulture, Sichuan Agricultural University","place":["Chengdu, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Guanjun","family":"Chen","sequence":"additional","affiliation":[{"name":"Tea Resources Utilization and Quality Testing Key Laboratory of Sichuan Province, Sichuan Agricultural University","place":["Chengdu, China"]},{"name":"College of Horticulture, Sichuan Agricultural University","place":["Chengdu, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jianzhao","family":"Wang","sequence":"additional","affiliation":[{"name":"Tea Resources Utilization and Quality Testing Key Laboratory of Sichuan Province, Sichuan Agricultural University","place":["Chengdu, China"]},{"name":"College of Horticulture, Sichuan Agricultural University","place":["Chengdu, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dandan","family":"Tang","sequence":"additional","affiliation":[{"name":"Tea Resources Utilization and Quality Testing Key Laboratory of Sichuan Province, Sichuan Agricultural University","place":["Chengdu, China"]},{"name":"College of Horticulture, Sichuan Agricultural University","place":["Chengdu, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Qinling","family":"Liu","sequence":"additional","affiliation":[{"name":"Tea Resources Utilization and Quality Testing Key Laboratory of Sichuan Province, Sichuan Agricultural University","place":["Chengdu, China"]},{"name":"College of Horticulture, Sichuan Agricultural University","place":["Chengdu, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiong","family":"Xiong","sequence":"additional","affiliation":[{"name":"College of Information Engineering, Sichuan Agricultural University","place":["Yucheng District, Ya\u2019an, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Liqiang","family":"Tan","sequence":"additional","affiliation":[{"name":"Tea Resources Utilization and Quality Testing Key Laboratory of Sichuan Province, Sichuan Agricultural University","place":["Chengdu, China"]},{"name":"College of Horticulture, Sichuan Agricultural University","place":["Chengdu, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,8,20]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","DOI":"10.3390\/horticulturae11111401","article-title":"Farmers\u2019 perceptions, preferences and choices of commercial tea clones in Uganda: From the smallholder out-growers\u2019 perspective","volume":"11","author":"Akankwasa","year":"2025","journal-title":"Horticulturae"},{"key":"B2","doi-asserted-by":"publisher","first-page":"14078","DOI":"10.1109\/access.2021.3051085","article-title":"Classification of remote sensing images using efficientnet-b3 cnn model with attention","volume":"9","author":"Alhichri","year":"2021","journal-title":"IEEE Access"},{"key":"B3","doi-asserted-by":"publisher","first-page":"20251281","DOI":"10.1515\/biol-2025-1281","article-title":"Assessment of nutrient deficiency of rice plant based on modified resnet50","volume":"21","author":"Behera","year":"2026","journal-title":"Open Life Sci."},{"key":"B4","doi-asserted-by":"publisher","DOI":"10.1016\/j.heliyon.2025.e42385","article-title":"Enhanced classification of medicinal plants using deep learning and optimized cnn architectures","volume":"11","author":"Bouakkaz","year":"2025","journal-title":"Heliyon"},{"key":"B5","doi-asserted-by":"publisher","DOI":"10.3389\/fpls.2023.1251418","article-title":"Oolong tea cultivars categorization and germination period classification based on multispectral information","volume":"14","author":"Cao","year":"2023","journal-title":"Front. Plant Sci."},{"key":"B6","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-642-31878-8","volume-title":"Global Tea Breeding: Achievements, Challenges and Perspectives","author":"Chen","year":"2012"},{"key":"B7","doi-asserted-by":"publisher","first-page":"588","DOI":"10.13305\/j.cnki.jts.2022.04.008","article-title":"Construction and application of an ecological-economic cultivar evaluation model for producing high-quality strip-shaped green tea","volume":"42","author":"Chen","year":"2022","journal-title":"J. Tea Sci."},{"key":"B8","doi-asserted-by":"publisher","DOI":"10.3390\/ijms232214169","article-title":"Effect of shading on the morphological, physiological, and biochemical characteristics as well as the transcriptome of matcha green tea","volume":"23","author":"Chen","year":"2022","journal-title":"Int. J. Mol. Sci."},{"key":"B9","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":"B10","doi-asserted-by":"publisher","DOI":"10.3390\/f14040728","article-title":"A non-destructive method for identification of tea plant cultivars based on deep learning","volume":"14","author":"Ding","year":"2023","journal-title":"Forests"},{"key":"B11","doi-asserted-by":"publisher","DOI":"10.1016\/j.indcrop.2024.118635","article-title":"Identification of varieties in camellia oleifera leaf based on deep learning technology","volume":"216","author":"Dong","year":"2024","journal-title":"Ind. Crops Prod."},{"key":"B12","doi-asserted-by":"publisher","first-page":"452","DOI":"10.1016\/j.tifs.2021.12.024","article-title":"A comprehensive review on microbiome, aromas and flavors, chemical composition, nutrition and future prospects of fuzhuan brick tea","volume":"119","author":"Du","year":"2022","journal-title":"Trends Food Sci. Technol."},{"key":"B13","doi-asserted-by":"publisher","first-page":"517","DOI":"10.1016\/j.foodchem.2016.09.047","article-title":"Variation of theanine, phenolic, and methylxanthine compounds in 21 cultivars of camellia sinensis harvested in different seasons","volume":"220","author":"Fang","year":"2017","journal-title":"Food Chem."},{"key":"B14","doi-asserted-by":"publisher","first-page":"469","DOI":"10.1007\/s00606-011-0559-3","article-title":"Genetic diversity and relationship of clonal tea (camellia sinensis) cultivars in China as revealed by ssr markers","volume":"298","author":"Fang","year":"2011","journal-title":"Plant Syst. Evol."},{"key":"B15","doi-asserted-by":"publisher","first-page":"14035","DOI":"10.1038\/hortres.2014.35","article-title":"Varietal identification of tea (camellia sinensis) using nanofluidic array of single nucleotide polymorphism (snp) markers","volume":"1","author":"Fang","year":"2014","journal-title":"Hortic. Res."},{"key":"B16","doi-asserted-by":"publisher","DOI":"10.3389\/fpls.2025.1662251","article-title":"Smart intercropping system to detect leaf disease using hyperspectral imaging and hybrid deep learning for precision agriculture","volume":"16","author":"Goyal","year":"2025","journal-title":"Front. Plant Sci."},{"key":"B17","doi-asserted-by":"publisher","first-page":"1343","DOI":"10.1111\/tpj.17053","article-title":"Deep learning to capture leaf shape in plant images: Validation by geometric morphometrics","volume":"120","author":"Hodac","year":"2024","journal-title":"Plant J."},{"key":"B18","doi-asserted-by":"publisher","first-page":"446","DOI":"10.38212\/2224-6614.3465","article-title":"Ssr-based molecular diagnosis for Taiwan tea cultivars and its application in identifying cultivar composition of the processed tea","volume":"31","author":"Hu","year":"2023","journal-title":"J. Food Drug Anal."},{"key":"B19","doi-asserted-by":"publisher","first-page":"68","DOI":"10.13305\/j.cnki.jts.2016.01.009","article-title":"Ssr identification and pedigree analysis of pvp application cultivars in tea plant","volume":"36","author":"Huang","year":"2016","journal-title":"J. Tea Sci."},{"key":"B20","doi-asserted-by":"publisher","DOI":"10.1186\/s13007-026-01518-5","article-title":"Coleaf, an image recognition-driven approach facilitates the genome-wide association study with tea leaf morphology","volume":"22","author":"Jiang","year":"2026","journal-title":"Plant Methods"},{"key":"B21","doi-asserted-by":"publisher","first-page":"4735","DOI":"10.1021\/acsomega.1c04102","article-title":"Identification of rice seed varieties based on near-infrared hyperspectral imaging technology combined with deep learning","volume":"7","author":"Jin","year":"2022","journal-title":"ACS Omega"},{"key":"B22","doi-asserted-by":"publisher","DOI":"10.1016\/j.compag.2020.105672","article-title":"A review of computer vision technologies for plant phenotyping","volume":"176","author":"Li","year":"2020","journal-title":"Comput. Electron. Agric."},{"key":"B23","doi-asserted-by":"publisher","first-page":"159","DOI":"10.25165\/j.ijabe.20221503.6896","article-title":"High-efficiency tea shoot detection method via a compressed deep learning model","volume":"15","author":"Li","year":"2022","journal-title":"Int. J. Agric. Biol. Eng."},{"key":"B24","doi-asserted-by":"publisher","DOI":"10.1016\/j.compag.2022.107352","article-title":"Leaf vein segmentation with self-supervision","volume":"203","author":"Li","year":"2022","journal-title":"Comput. Electron. Agric."},{"key":"B25","doi-asserted-by":"publisher","DOI":"10.3389\/fpls.2024.1442225","article-title":"Advances in the tea plants phenotyping using hyperspectral imaging technology","volume":"15","author":"Luo","year":"2024","journal-title":"Front. Plant Sci."},{"key":"B26","doi-asserted-by":"publisher","DOI":"10.3389\/fpls.2026.1789542","article-title":"Explainable artificial-intelligence-based hyperspectral image analysis for leaf disease detection in intercropping system","volume":"17","author":"Malik","year":"2026","journal-title":"Front. Plant Sci."},{"key":"B27","doi-asserted-by":"publisher","first-page":"10","DOI":"10.20425\/ijts.v13i01-02.9981","article-title":"Seasonal and environment variations of yields and yield components of tea (camellia sinensis) cultivars in Kenya","volume":"13","author":"Nyabundi","year":"2017","journal-title":"Int. J. Tea Sci."},{"key":"B28","doi-asserted-by":"publisher","DOI":"10.1007\/s11760-025-04445-5","article-title":"A vision transformer ensemble and mobile augmented reality solution for mushroom toxicity classification","volume":"19","author":"\u00d6zben","year":"2025","journal-title":"Signal. Image Video Process."},{"key":"B29","doi-asserted-by":"publisher","first-page":"69","DOI":"10.1016\/j.compag.2018.08.001","article-title":"Computer vision and artificial intelligence in precision agriculture for grain crops: A systematic review","volume":"153","author":"Patr\u00edcio","year":"2018","journal-title":"Comput. Electron. Agric."},{"key":"B30","doi-asserted-by":"publisher","first-page":"19249","DOI":"10.1109\/access.2026.3656107","article-title":"Modality-aware dual-stream biometric authentication using convnext-tiny and tinyvit with attention-based feature fusion","volume":"14","author":"Sajid","year":"2026","journal-title":"IEEE Access"},{"key":"B31","doi-asserted-by":"publisher","first-page":"223","DOI":"10.6041\/j.issn.1000-1298.2023.02.022","article-title":"Classification method of multi-variety tea leaves based on improved squeezenet model","volume":"54","author":"Sun","year":"2023","journal-title":"Trans. Chin. Soc. For. Agric. Machinery"},{"key":"B32","doi-asserted-by":"publisher","DOI":"10.3390\/ijms20246196","article-title":"Health functions and related molecular mechanisms of tea components: An update review","volume":"20","author":"Tang","year":"2019","journal-title":"Int. J. Mol. Sci."},{"key":"B33","doi-asserted-by":"publisher","DOI":"10.3390\/electronics12102285","article-title":"An efficient classification of rice variety with quantized neural networks","volume":"12","author":"Tasci","year":"2023","journal-title":"Electronics"},{"key":"B34","article-title":"Guidelines for the Conduct of Tests for Distinctness, Uniformity and Stability","year":"2022"},{"key":"B35","doi-asserted-by":"publisher","first-page":"77","DOI":"10.13305\/j.cnki.jts.2016.01.010","article-title":"Leaf functional and photosynthetic characteristics in different leaves positions of tea plant","volume":"36","author":"Wang","year":"2016","journal-title":"J. Tea Sci."},{"key":"B36","doi-asserted-by":"publisher","first-page":"1152","DOI":"10.1186\/s40064-016-2835-8","article-title":"An efficient identification strategy of clonal tea cultivars using long-core motif ssr markers","volume":"5","author":"Wang","year":"2016","journal-title":"Springerplus"},{"key":"B37","doi-asserted-by":"publisher","DOI":"10.3389\/fpls.2025.1697209","article-title":"Ecs-tea: A bio-inspired high-precision detection and localization algorithm for young shoots of pu-erh tea","volume":"16","author":"Wang","year":"2025","journal-title":"Front. Plant Sci."},{"key":"B38","doi-asserted-by":"publisher","DOI":"10.3390\/sym13101769","article-title":"Deep neural network compression for plant disease recognition","volume":"13","author":"Wang","year":"2021","journal-title":"Symmetry"},{"key":"B39","doi-asserted-by":"publisher","DOI":"10.3389\/fpls.2024.1382802","article-title":"Tomato leaf disease detection based on attention mechanism and multi-scale feature fusion","volume":"15","author":"Wang","year":"2024","journal-title":"Front. Plant Sci."},{"key":"B40","doi-asserted-by":"publisher","DOI":"10.1016\/j.infrared.2022.104533","article-title":"A recognition method of corn varieties based on spectral technology and deep learning model","volume":"128","author":"Yang","year":"2023","journal-title":"Infrared Phys. Technol."},{"key":"B41","doi-asserted-by":"publisher","DOI":"10.1016\/j.sciaf.2025.e03061","article-title":"Mango fruit disease detection and classification using mobilenetv3_large model","volume":"30","author":"Yehulu","year":"2025","journal-title":"Sci. Afr."},{"key":"B42","doi-asserted-by":"publisher","first-page":"563","DOI":"10.1016\/j.hpj.2022.08.007","article-title":"Natural variation of main biochemical components, morphological and yield traits among a panel of 87 tea [camellia sinensis (l.) o. Kuntze] cultivars","volume":"9","author":"Zaman","year":"2023","journal-title":"Hortic. Plant J."},{"key":"B43","doi-asserted-by":"publisher","DOI":"10.3389\/fpls.2020.00280","article-title":"Screening tea cultivars for novel climates: Plant growth and leaf quality of camellia sinensis cultivars grown in mississippi, United States","volume":"11","author":"Zhang","year":"2020","journal-title":"Front. Plant Sci."},{"key":"B44","doi-asserted-by":"publisher","DOI":"10.3390\/agriculture15192027","article-title":"Training strategy optimization of a tea canopy dataset for variety identification during the harvest period","volume":"15","author":"Zhang","year":"2025","journal-title":"Agriculture"},{"key":"B45","doi-asserted-by":"publisher","DOI":"10.1016\/j.compag.2025.111140","article-title":"A hybrid and high-performance model for tea cultivars discrimination using canopy images with convolutional neural network and machine learning algorithm","volume":"240","author":"Zhang","year":"2026","journal-title":"Comput. Electron. Agric."},{"key":"B46","doi-asserted-by":"publisher","DOI":"10.1016\/j.scienta.2024.113908","article-title":"Identification of tea plant cultivars based on canopy images using deep learning methods","volume":"339","author":"Zhang","year":"2025","journal-title":"Sci. Hortic."},{"key":"B47","doi-asserted-by":"publisher","DOI":"10.48130\/bpr-0024-0012","article-title":"Metabolome profiling unveil the composition differences of quality of different tea cultivars","volume":"4","author":"Zhao","year":"2024","journal-title":"Beverage Plant Res."},{"key":"B48","doi-asserted-by":"publisher","DOI":"10.1016\/j.compag.2022.106703","article-title":"Identification method of vegetable diseases based on transfer learning and attention mechanism","volume":"193","author":"Zhao","year":"2022","journal-title":"Comput. Electron. Agric."}],"container-title":["Frontiers in Plant Science"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpls.2026.1863075\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,20]],"date-time":"2026-08-20T05:28:34Z","timestamp":1787203714000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpls.2026.1863075\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,20]]},"references-count":48,"alternative-id":["10.3389\/fpls.2026.1863075"],"URL":"https:\/\/doi.org\/10.3389\/fpls.2026.1863075","relation":{},"ISSN":["1664-462X"],"issn-type":[{"value":"1664-462X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,20]]},"article-number":"1863075"}}