High-throughput seed phenotyping of Populus cultivars in China using vibration-assisted machine vision with alternating back-lit and front-lit illuminations

Authors

  • Xiwei Wang 1. College of Mechanical and Electronic Engineering, Nanjing Forestry University, Nanjing 210037, China
  • Mubikayi Muhong Horly 1. College of Mechanical and Electronic Engineering, Nanjing Forestry University, Nanjing 210037, China
  • Zanpeng Li 1. College of Mechanical and Electronic Engineering, Nanjing Forestry University, Nanjing 210037, China
  • Maocheng Zhao 1. College of Mechanical and Electronic Engineering, Nanjing Forestry University, Nanjing 210037, China
  • Bin Wu 1. College of Mechanical and Electronic Engineering, Nanjing Forestry University, Nanjing 210037, China
  • Mengmeng Wang 1. College of Mechanical and Electronic Engineering, Nanjing Forestry University, Nanjing 210037, China
  • Jiale Deng 1. College of Mechanical and Electronic Engineering, Nanjing Forestry University, Nanjing 210037, China
  • Mengmeng Qiao 1. College of Mechanical and Electronic Engineering, Nanjing Forestry University, Nanjing 210037, China
  • Xiao Chen 1. College of Mechanical and Electronic Engineering, Nanjing Forestry University, Nanjing 210037, China
  • Yongjie Lu 1. College of Mechanical and Electronic Engineering, Nanjing Forestry University, Nanjing 210037, China
  • Liang Qi 1. College of Mechanical and Electronic Engineering, Nanjing Forestry University, Nanjing 210037, China
  • Weijun Xie 1. College of Mechanical and Electronic Engineering, Nanjing Forestry University, Nanjing 210037, China
  • Hongyuan Zou 1. College of Mechanical and Electronic Engineering, Nanjing Forestry University, Nanjing 210037, China
  • Yan Zhang 2. Liaoning Institute of Poplar Research, Gaizhou 115213, Liaoning, China
  • Rusheng Peng 2. Liaoning Institute of Poplar Research, Gaizhou 115213, Liaoning, China

Keywords:

Populus seed, machine vision, camera calibration, flexible vibratory plate, ortho-image.

Abstract

Seeds of major Populus cultivars were collected from across China in 2024 to build the image-data bank of over <styled-content style-type="number">1187000</styled-content> images of singular seeds for the National Forestry and Grassland Science Data Center (NFGSDC). An innovative vibration-assisted machine-vision system was built with alternating back-lit and front-lit illumination, which incorporated a flexible vibratory panel (FVP) to manipulate the multitude of seeds to minimize the occurrence of butting or overlapping, and the lighting from alternating directions to capture phenotypic features both in silhouettes and in vivid color images. To investigate how illumination directions would affect phenotyping, morphological and chromatic metrics were measured, respectively from only the common front-lit images and through the combined use with back-lit images, and applied to distinguish different cultivars and harvest-batches. Results verified that back-lit excelled for reliable segmentation for feature images and accurate morphological metrics, especially when the closeness was clearly revealed in the clustering dendrogram between Nanlin 895 and Zhonglin 46, which shared a common genetic sourcing from P. Euramericana. In contrast, front-lit images were prone to occasional segmentation defects leading to inaccurate morphological measurements due to the highly dynamic range of seed colors, which caused the clustering to lose the genetic relevance. The power of the image-dataset of alternating illuminations was further demonstrated when a decent accuracy of 0.819 yielded from the simple support-vector-machine classification while working on only the back-lit morphological measurements, and the increase to 0.856 with statistical significance if with the addition of chromatic metrics from corresponding front-lit color images, while other image characteristics had been strictly held back. The vibration-assisted alternating illumination protocol established in this work to capture delicate seed-features of Populus cultivars may also be applied to other small grains facing similar imaging challenges, laying a sturdy step-stone of high-throughput phenotyping for large-scale breeding programs and genetic studies.      

Keywords: Populus seed, machine vision, camera calibration, flexible vibratory plate, back-lit and front-lit illumination

DOI: 10.25165/j.ijabe.20261901.9850

Citation: Wang X W, Horly M M, Li Z P, Zhao M C, Wu B, Wang M M, et al. High-throughput seed phenotyping of Populus cultivars in China using vibration-assisted machine vision with alternating back-lit and front-lit illuminations. Int J Agric & Biol Eng, 2026; 19(1): 197–212.

References

[1] Yun X, Chen Y. International development of saline-alkali land and its enlightenment to China. Territory & Natural Resources Study, 2020; 1: 84–87.

[2] Zhang Q, Huang J, Yang J, Guan X, Yu H, Zhu B, et al. Advances in research on climate change and its effects on the arid and semi-arid regions of China over the past century. J Meteorol Res, 2025; 39（3）: 673–687.

[3] Yang Z L, Zhou S Y, Zhang W D, Yang Z X. Poplar genetic resources in North China: the challenge of sustainable forestry. Forest Genetic Resources （FAO）, 1999; No. 27. https://www.fao.org/4/x4133e/X4133E02.htm

[4] Liu S, Huang Z, Xu Z, Zhao F, Xiong D, Peng S, et al. High-throughput measurement method for rice seedling based on improved UNet model. Computers and Electronics in Agriculture, 2024; 219: 108770.

[5] Tu K, Wu W, Cheng Y, Zhang H, Xu Y, Dong X, et al. AIseed: An automated image analysis software for high-throughput phenotyping and quality non-destructive testing of individual plant seeds. Computers and Electronics in Agriculture, 2023; 207: 107740.

[6] Baek J, Lee E, Kim N, Kim S L, Choi I, Ji H, et al. High throughput phenotyping for various traits on soybean seeds using image analysis. Sensors, 2020; 20（1）: 248.

[7] Van De Looverbosch T, Vandenbussche B, Verboven P, Nicolaï B. Nondestructive high-throughput sugar beet fruit analysis using X-ray CT and deep learning. Computers and Electronics in Agriculture, 2022; 200: 107228.

[8] Huang X, Zheng S, Zhu N. High-throughput legume seed phenotyping using a handheld 3D laser scanner. Remote Sensing, 2022; 14（2）: 431.

[9] Lin T Y, Huang P C, Syu C Y, Huang K Y. Automatic inspection system for red paddy seeds. Sensors and Materials, 2023; 35（12）: 4383–4395.

[10] Ozkaya Y A. Digital image processing and illumination techniques for yarn characterization. J Electron Imaging, 2005; 14（2）: 023001.

[11] Zu Q, Liu T, Zhu W, Pan Y, Wang J, Song X, et al. Automated seed counting using image processing and deep learning. Front Plant Sci., 2025; 16: 1659781.

[12] Loddo A, Loddo M, Di Ruberto C. A novel deep learning based approach for seed image classification and retrieval. Computers and Electronics in Agriculture, 2021; 187: 106269.

[13] Mishra P, Lohumi S, Ahmad Khan H, Nordon A. Close-range hyperspectral imaging of whole plants for digital phenotyping: Recent applications and illumination correction approaches. Computers and Electronics in Agriculture, 2020; 178: 105780.

[14] Chopin J, Kumar P, Miklavcic S J. Land-based crop phenotyping by image analysis: consistent canopy characterization from inconsistent field illumination. Plant Methods, 2018; 14（1）: 39.

[15] Dayrell R L C, Ott T, Horrocks T, Poschlod P. Automated extraction of seed morphological traits from images. Methods Ecol Evol, 2023; 14（7）: 1708–18.

[16] Halcro K, McNabb K, Lockinger A, Socquet-Juglard D, Bett K E, Noble S D. The BELT and phenoSEED platforms: shape and colour phenotyping of seed samples. Plant Methods, 2020; 16（1）: 49.

[17] National Science & Technology Infrastructure of China, National Forestry and Grassland Science Data Center （NFGSDC）. Singular seed visible image dataset of poplar species from eastern, northeastern, northern, and central China. 2025; Available: https://www.forestdata.cn/dataDetail.html?id=aed290f4-60dd-4812-bfe6-53d396559846

[18] National Science & Technology Infrastructure of China, National Forestry and Grassland Science Data Center （NFGSDC）. Seeds on tray under back-light and front-light illumination visible image dataset of poplar species from eastern, northeastern, northern, and central China. 2025; Available: https://www.forestdata.cn/dataDetail.html?id=01946496-999a-4883-89d1-b7db3335b84e

[19] Liu X, Hou L, Ding C, Su X, Zhang W, Pang Z, et al. Effects of stand age and soil microbial communities on soil respiration throughout the growth cycle of poplar plantations in northeastern China. Front Microbiol, 2024; 15: 1477571.

[20] Smreciu A, Landhäusser S, Marenholtz E, Sobze J M, Gould K, Schoonmaker A. Aspen seed collection and cleaning. 2013; Available: https://www.cclmportal.ca/sites/default/files/2020-02/Aspen_Seed_Collection_and_Cleaning.pdf

[21] Brown K R. Catkin growth, seed production, and development of seed germinability in quaking aspen in central Alberta. Tree Planters’ Notes, 1989; 40（2）: 25–29. https://digitalcommons.usu.edu/aspen_bib3209

[22] Fung M Y P, Hamel B A. Aspen seed collection and extraction. Tree Planters’ Notes, 1993; 44（3）: 98–100.

[23] Johnson L. Effect of humidity on the longevity of Populus and Ulmus seeds in storage. Canadian Journal of Research, 1946; 24（6）: 298–302.

[24] Maisenhelder L C. Planting and growing cottonwood on bottomlands. State College, Miss.: Mississipp.Agricultural Experiment Station, 1951. Available: https://catalog.hathitrust.org/Record/011461492

[25] Cui B X, Liu J L, Zhou N, Wu W P, Tang X H, Ding C J, et al. Effect of thinning intensity on fiber morphology and crystallinity of poplar. BioRes 2024; 20（1）: 1024–1036.

[26] Bhutta N, Nunez-Martinez O F, Mei C, Bräutigam K. Seed collection in temperate trees—clean, fast, and effective extraction of populus seeds for laboratory use and long-term storage. Bio-protocol, 2024; 14（3）: e4927.

[27] Einspahr D, Schlafke D. A method for Aspen and Cottonwood seed extraction. Tree Planters’ Notes, 1957; 28: 10.

[28] Roe E I, McCain D P. A quick method of collecting and cleaning Aspen seed. Tree Planters’ Notes, 1961; 51: 17–18.

[29] Liu Y, Wu X, Zhang J, Liu S, Semple K, Dai C. Maturation stress and sood properties of poplar （Populus × euramericana cv. ‘Zhonglin46’） tension wood. Forests, 2023; 14（7）: 1505.

[30] Weisgerber H, Han Y. Diversity and breeding potential of poplar species in China. The Forestry Chronicle, 2001; 77（2）: 227–237.

[31] Borkhert E V, Pushkova E N, Nasimovich Y A, Kostina M V, Vasilieva N V, Murataev R A, et al. Sex-determining region complements traditionally used in phylogenetic studies nuclear and chloroplast sequences in investigation of Aigeiros Duby and Tacamahaca Spach poplars （genus Populus L. , Salicaceae）. Front Plant Sci, 2023; 14: 1204899.

[32] Faust M E. Germination of Populus grandidentata and P. tremuloides, with particular reference to oxygen consumption. Botanical Gazette, 1936; 97（4）: 808–821. Available: http://www.jstor.org/stable/2471529

[33] Junaidi A, Andriyanto M. Seed collection time effect on the germination rate and growth of rubber tree rootstock. In: Proceedings of the 3rd KOBI Congress, International and National Conferences （KOBICINC 2020）. Atlantis Press; 2021; pp.278–282. doi: 10.2991/absr.k.210621.046

[34] Zafari S, Eerola T, Sampo J, Kalviainen H, Haario H. Segmentation of overlapping elliptical objects in silhouette images. IEEE Trans on Image Process, 2015; 24（12）: 5942–5952.

[35] Han L, Deng J. A study on flexible vibratory feeding system based on HALCON machine vision software. Zhengzhou, China, 2015. Available: https://www.atlantis-press.com/article/18191 Accessed on [2024-09-16]

[36] Rao S S. Vibration of continuous systems. 1st ed. Wiley; 2019. Available: https://onlinelibrary.wiley.com/doi/book/10.1002/9781119424284. Accessed on [2025-11-20]

[37] Steinecke C, Lee J, Friedman J. A standardized and efficient technique to estimate seed traits in plants with numerous small propagules. App. Plant Sci., 2023; 11（5）: e11552.

[38] Ghimire A, Kim S H, Cho A, Jang N, Ahn S, Islam M S, et al. Automatic evaluation of soybean seed traits using RGB image data and a python algorithm. Plants, 2023; 12（17）: 3078.

[39] Dayrell R L C, Ott T, Horrocks T, Poschlod P. Automated extraction of seed morphological traits from images. Methods Ecol Evol, 2023; 14（7）: 1708–1718.

[40] Cervantes E, Martín J J, Saadaoui E. Updated methods for seed shape analysis. Scientifica, 2016; 2016: 1–10.

[41] Florczyk S. Video based indoor exploration with autonomous and mobile robots. Journal of Intelligent and Robotic Systems, 2005; 41（4）: 245–262.

[42] Baek J, Lee E, Kim N, Kim S L, Choi I, Ji H, et al. High throughput phenotyping for various traits on soybean seeds using image analysis. Sensors, 2020; 20（1）: 248.

[43] Tu K, Wu W, Cheng Y, Zhang H, Xu Y, Dong X, et al. AIseed: An automated image analysis software for high-throughput phenotyping and quality non-destructive testing of individual plant seeds. Computers and Electronics in Agriculture, 2023; 207: 107740.

[44] Hernández S, Zhong V, Brophy J A N. SeedSeg: image-based transgenic seed counting for segregation analysis of T-DNA loci. Plant Methods, 2025; 21（1）: 87.

[45] Demšar J. Statistical comparisons of classifiers over multiple data sets. Journal of Machine Learning Research, 2006; 7（1）: 1–30. Available: https://www.jmlr.org/papers/volume7/demsar06a/demsar06a.pdf

[46] Barrozo M A S, Mujumdar A, Freire J T. Air-drying of seeds: A review. Drying Technology, 2014; 32（10）: 1127–1141.

[47] Corbineau F. The effects of storage conditions on seed deterioration and ageing: How to improve seed longevity. Seeds, 2024; 3（1）: 56–75.

[48] Silva J H C S, Azerêdo G A D, Souza V C D. Conservation of seeds of cactaceae species endemic to the caatinga biome: Pilosocereus pachycladus and Tacinga inamoena. Rev Caatinga, 2023; 36（1）: 115–123.

[49] Tiebel K, Dahlmann J, Karge A. Global warming could shorten the seed lifespan of pioneer tree species and thus natural regeneration window of damaged areas. European Journal of Forest Research, 2024; 143（2）: 437–450.

[50] Wyckoff G, Zasada J C. Populus L, Bonner F. Woody plant seed manual. Agric Handbook, 2005; 727: 856–871. Available: https://www.fs.usda.gov/nsl/Wpsm/Populus.pdf

[51] Young J A, Young C G. Seeds of woody plants in North America. Rev. and enl. ed. Portland, Or: Dioscorides Press; 1992.

[52] Xu W, Qi H, Shen T, Zhao M, Song Z, Ran N, et al. Poplar coma morphogenesis and miRNA regulatory networks by combining ovary tissue sectioning and deep sequencing. iScience, 2023; 26（4）: 106496.

[53] Liu W, Liu C, Jin J, Li D, Fu Y, Yuan X. High-throughput phenotyping of morphological seed and fruit characteristics using X-ray computed tomography. Front Plant Sci, 2020; 11: 601475.

[54] Han L, Wang L Y, Hu G P. A study on the machine vision assisted vibratory feeding system. Applied Mechanics and Materials, 2012; 220: 1377–1380.

[55] Han L, Deng J. A study on flexible vibratory feeding system based on halcon machine vision software. Atlantis Press, 2015; pp.9–12. Available: https://www.atlantis-press.com/article/18191

[56] Wu D, Ding D, Cui B, Jiang S, Zhao E, Liu Y, et al. Design and experiment of vibration plate type camellia fruit picking machine. Int J Agric & Biol Eng, 2022; 15（4）: 130–138.

[57] Ropelewska E, Sabanci K, Aslan M F, Azizi A. A novel approach to the authentication of apricot seed cultivars using innovative models based on image texture parameters. Horticulturae, 2022; 8（5）: 431.

[58] Ropelewska E. The use of seed texture features for discriminating different cultivars of stored apples. Journal of Stored Products Research, 2020; 88: 101668.

[59] Medina W, Skurtys O, Aguilera J M. Study on image analysis application for identification Quinoa seeds （Chenopodium quinoa Willd） geographical provenance. LWT-Food Science and Technology, 2010; 43（2）: 238–246.

[60] Sharma K K, Seal A. Clustering analysis using an adaptive fused distance. Engineering Applications of Artificial Intelligence, 2020; 96: 103928.

[61] Van De Velden M, Iodice D’Enza A, Markos A. Distance‐based clustering of mixed data. WIREs Computational Stats, 2019; 11（3）: e1456.

[62] Deok Han G, Mansoor S, Kim J, Park J, Heo S, Yu J K, et al. A study of the morphological and geographical diversity of Korean indigenous buckwheat landraces for breeding. Journal of King Saud University-Science, 2024; 36（9）: 103387.

[63] Juma I, Nyomora A, Hovmalm H P, Fatih M, Geleta M, Carlsson AS, et al. Characterization of Tanzanian avocado using morphological traits. Diversity, 2020; 12（2）: 64.

[64] Han D, Hu H, Yang J, Liang X, Ai J, Abula A, et al. The ideal harvest time for seed production in maize （Zea mays L.） varieties of different maturity groups. J Sci Food Agric, 2022; 102（13）: 5867–5874.

[65] Morales C G D P, García-De Los Santos G, Aguilar-Rincón V H, Hernández-Livera A, Escamilla-Prado E. Coffee （Coffea arabica L.） harvesting time and its influence on the seed quality of the Costa Rica 95 and Garnica varieties. Agro Productividad, 2022; DOI: 10.32854/agrop.v15i8.1935.

Downloads

Published

2026-03-16

How to Cite

(1)
Wang, X.; Horly, M. M.; Li, Z.; Zhao, M.; Wu, B.; Wang, M.; Deng, J.; Qiao, M.; Chen, X.; Lu, Y. High-Throughput Seed Phenotyping of Populus Cultivars in China Using Vibration-Assisted Machine Vision With Alternating Back-Lit and Front-Lit Illuminations. Int J Agric &amp; Biol Eng 2026, 19.

Issue

Vol. 19 No. 1 (2026): IJABE

Section

Information Technology, Sensors and Control Systems

License

Copyright (c) 2026 International Journal of Agricultural and Biological Engineering

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

IJABE is an international peer reviewed, open access journal, adopting Creative Commons Copyright Notices as follows.

Authors who publish with this journal agree to the following terms:

  1. Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
  2. Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
  3. Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).