[en] Throughout their lifetime, trees store valuable environmental information within their wood. Unlocking this information requires quantitative analysis, in most cases of the surface of wood. The conventional pathway for high-resolution digitization of wood surfaces and segmentation of wood features requires several manual and time consuming steps. We present a semi-automated high-throughput pipeline for sample preparation, gigapixel imaging, and analysis of the anatomy of the end-grain surfaces of discs and increment cores. The pipeline consists of a collaborative robot (Cobot) with sander for surface preparation, a custom-built open-source robot for gigapixel imaging (Gigapixel Woodbot), and a Python routine for deep-learning analysis of gigapixel images. The robotic sander allows to obtain high-quality surfaces with minimal sanding or polishing artefacts. It is designed for precise and consistent sanding and polishing of wood surfaces, revealing detailed wood anatomical structures by applying consecutively finer grits of sandpaper. Multiple samples can be processed autonomously at once. The custom-built open-source Gigapixel Woodbot is a modular imaging system that enables automated scanning of large wood surfaces. The frame of the robot is a CNC (Computer Numerical Control) machine to position a camera above the objects. Images are taken at different focus points, with a small overlap between consecutive images in the X-Y plane, and merged by mosaic stitching, into a gigapixel image. Multiple scans can be initiated through the graphical application, allowing the system to autonomously image several objects and large surfaces. Finally, a Python routine using a trained YOLOv8 deep learning network allows for fully automated analysis of the gigapixel images, here shown as a proof-of-concept for the quantification of vessels and rays on full disc surfaces and increment cores. We present fully digitized beech discs of 30-35 cm diameter at a resolution of 2.25 μ m, for which we automatically quantified the number of vessels (up to 13 million) and rays. We showcase the same process for five 30 cm length beech increment cores also digitized at a resolution of 2.25 μ m, and generated pith-to-bark profiles of vessel density. This pipeline allows researchers to perform high-detail analysis of anatomical features on large surfaces, test fundamental hypotheses in ecophysiology, ecology, dendroclimatology, and many more with sufficient sample replication.
Disciplines :
Environmental sciences & ecology
Author, co-author :
Van den Bulcke, Jan; UGent-Woodlab, Department of Environment, Ghent University, Coupure Links 653, 9000, Gent, Belgium. Jan.VandenBulcke@UGent.be
Verschuren, Louis; UGent-Woodlab, Department of Environment, Ghent University, Coupure Links 653, 9000, Gent, Belgium ; ForNaLab, Department of Environment, Ghent University, Geraardsbergsesteenweg 267, 9090, Merelbeke (Gontrode), Belgium
De Blaere, Ruben; UGent-Woodlab, Department of Environment, Ghent University, Coupure Links 653, 9000, Gent, Belgium ; Service of Wood Biology, Royal Museum for Central Africa, Leuvensesteenweg 13, 3080, Tervuren, Belgium
Vansuyt, Simon; UGent-Woodlab, Department of Environment, Ghent University, Coupure Links 653, 9000, Gent, Belgium ; AI and Robotics Lab, IDLab-AIRO, Ghent University - imec, Technologiepark-Zwijnaarde 126, 9000, Gent, Belgium
Dekegeleer, Maxime; UGent-Woodlab, Department of Environment, Ghent University, Coupure Links 653, 9000, Gent, Belgium
Kibleur, Pierre; UGent-Woodlab, Department of Environment, Ghent University, Coupure Links 653, 9000, Gent, Belgium ; RP-UGCT, Department of Physics and Astronomy - Radiation Physics, Ghent University, Proeftuinstraat 86, 9000, Gent, Belgium
Pieters, Olivier; AI and Robotics Lab, IDLab-AIRO, Ghent University - imec, Technologiepark-Zwijnaarde 126, 9000, Gent, Belgium
De Mil, Tom ; Université de Liège - ULiège > TERRA Research Centre > Gestion des ressources forestières
Hubau, Wannes; UGent-Woodlab, Department of Environment, Ghent University, Coupure Links 653, 9000, Gent, Belgium ; Service of Wood Biology, Royal Museum for Central Africa, Leuvensesteenweg 13, 3080, Tervuren, Belgium
Beeckman, Hans; Service of Wood Biology, Royal Museum for Central Africa, Leuvensesteenweg 13, 3080, Tervuren, Belgium
Van Acker, Joris; UGent-Woodlab, Department of Environment, Ghent University, Coupure Links 653, 9000, Gent, Belgium
Wyffels, Francis; AI and Robotics Lab, IDLab-AIRO, Ghent University - imec, Technologiepark-Zwijnaarde 126, 9000, Gent, Belgium
Language :
English
Title :
Enabling high-throughput quantitative wood anatomy through a dedicated pipeline.
UGent - Ghent University FWO - Fonds Wetenschappelijk Onderzoek Vlaanderen BELSPO - Belgian Federal Science Policy Office
Funding text :
This work was supported by the UGent Special Research Fund (BOF Starting Grant BOFSTG2018000701), the Research Foundation Flanders (FWO) through the ACTREAL (G019521N) and XyloDynaCT (G009720N) projects and the Belgian Science Policy Office through the BRAIN-be 2.0 program (Belgian Research Action through Interdisciplinary Networks PHASE 2-2018-2023) project SmartWoodID (B2/202/P2/SmartWoodID) and ENFORCE (RT/22/ENFORCE).
D. Griffin S.T. Porter M.L. Trumper K.E. Carlson D.J. Crawford D. Schwalen et al. Gigapixel macro photography of tree rings Tree-Ring Res 77 2 86 94 10.3959/TRR2021-3
R. De Blaere K. Lievens D. Van Hassel V. Deklerck T. De Mil W. Hubau et al. SmartWoodID-an image collection of large end-grain surfaces to support wood identification systems Database 2023 baad034 10.1093/database/baad034 37178209 10182821
J. Tardif F. Conciatori E. Yeung C. Stasolla M. Sumner B. Huang Microscopic examination of wood Sample preparation and techniques for light microscopy Cham Springer 331 44
H. Spiecker M.G. Schinker J. Hansen Y. Park T. Ebding W. Döll Cell structure in tree rings: Novel methods for preparation and image analysis of large cross sections IAWA Journal 21 3 361 73 10.1163/22941932-90000253
G. Von Arx A. Crivellaro A.L. Prendin K. Čufar M. Carrer Quantitative wood anatomy-practical guidelines Front Plant Sci 7 781
S. Fukuta M. Nomura T. Ikeda et al. UV laser machining of wood Eur J Wood Wood Prod 74 261 267 1:CAS:528:DC%2BC28XisFSntbk%3D 10.1007/s00107-016-1010-9
X. Guo M. Deng Y. Hu Y. Wang T. Ye Morphology, mechanism and kerf variation during CO2 laser cutting pine wood J Manufact Process 68 13 22 10.1016/j.jmapro.2021.05.036
A. Arzac J.M. López-Cepero E.A. Babushkina S. Gomez Applying methods of hard tissues preparation for wood anatomy: Imaging polished samples embedded in polymethylmethacrylate Dendrochronologia 51 76 81 10.1016/j.dendro.2018.08.005
F. Nagata Y. Kusumoto Y. Fujimoto K. Watanabe Robotic sanding system for new designed furniture with free-formed surface Robot Comput Integrat Manufact 23 4 371 9 10.1016/j.rcim.2006.04.004
S. Arenas-Castro J. Fernández-Haeger D. Jordano-Barbudo A method for tree-ring analysis using DIVA-GIS freeware on scanned core images Tree-ring Res 71 2 118 29 10.3959/1536-1098-71.2.118
M. Rydval L.Å. Larsson L. McGlynn B.E. Gunnarson N.J. Loader G.H. Young et al. Blue intensity for dendroclimatology: should we have the blues? Exp Scotland Dendrochronologia 32 3 191 204 10.1016/j.dendro.2014.04.003
K. Seftigen M.V. Fonti B. Luckman M. Rydval P. Stridbeck G. Von Arx et al. Prospects for dendroanatomy in paleoclimatology-a case study on Picea engelmannii from the Canadian Rockies Clim Past Discuss 2022 1 32
García-Hidalgo M, García-Pedrero Á, Colón D, Sangüesa-Barreda G, García-Cervigón A, López-Molina J, et al. CaptuRING: a do-it-yourself tool for wood sample digitization. Methods Ecol Evol. 2022; 1–7.
T. Levanič ATRICS-a new system for image acquisition in dendrochronology Tree-Ring Res 63 2 117 22 10.3959/1536-1098-63.2.117
S. Klesse T. Wohlgemuth K. Meusburger Y. Vitasse G. von Arx M. Lévesque et al. Long-term soil water limitation and previous tree vigor drive local variability of drought-induced crown dieback in Fagus sylvatica Sci Total Environ 851 157926 1:CAS:528:DC%2BB38Xit1Slu7jM 10.1016/j.scitotenv.2022.157926 35985592
J.C. Cerre Incident light microphotography at high depth of focus IAWA Journal 37 3 506 10 10.1163/22941932-20160149
M. Rydval J. Björklund G. von Arx K. Begović M. Lexa J. Nogueira et al. Ultra-high-resolution reflected-light imaging for dendrochronology Dendrochronologia 83 126160 10.1016/j.dendro.2023.126160
W. Liang I. Heinrich G. Helle I.D. Liñán T. Heinken Applying CLSM to increment core surfaces for histometric analyses: a novel advance in quantitative wood anatomy Dendrochronologia 31 2 140 5 10.1016/j.dendro.2012.09.002
W. Huang Y. Li Practical guidelines for quantitative wood anatomy on Ginkgo biloba L IAWA Journal 1(aop) 1 20
J. Van den Bulcke J. Dhaene V. Denis L. Hoorebeke M. Boone M. Boone et al. Advanced X-ray CT scanning can boost tree-ring research for earth-system sciences Ann Bot 124 837 847 10.1093/aob/mcz126 31361809 6868372
P. Fonti N. Solomonoff I. García-González Earlywood vessels of Castanea sativa record temperature before their formation New Phytol 173 3 562 70 10.1111/j.1469-8137.2006.01945.x 17244050
Kirillov A, Mintun E, Ravi N, Mao H, Rolland C, Gustafson L, et al. Segment anything. In: Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV). 2023;4015–4026.
Wu Y, Kirillov A, Massa F, Lo WY, Girshick R.: Detectron2. https://github.com/facebookresearch/detectron2.
Jocher G, Chaurasia A, Qiu J.: Ultralytics YOLO. https://github.com/ultralytics/ultralytics.
A. Fabijańska M. Danek DeepDendro—a tree rings detector based on a deep convolutional neural network Comput Electron Agric 150 353 63 10.1016/j.compag.2018.05.005
D. Kim C. Ko D. Kim Method for detecting tree ring boundary in conifers and broadleaf trees using Mask R-CNN and linear interpolation Dendrochronologia 79 126088 10.1016/j.dendro.2023.126088
M. Poláek A. Arizpe P. Hüther L. Weidlich S. Steindl K. Swarts Automation of tree-ring detection and measurements using deep learning Methods Ecol Evol 14 9 2233 42 10.1111/2041-210X.14183
G. Resente A. Gillert M. Trouillier A. Anadon-Rosell R.L. Peters G. von Arx et al. Mask, train, repeat! artificial intelligence for quantitative wood anatomy Front Plant Sci 10.3389/fpls.2021.767400 34804101 8601631
H. Ergun Segmentation of wood cell in cross-section using deep convolutional neural networks J Intell Fuzzy Syst 41 6 7447 56 10.3233/JIFS-211386
M. Katzenmaier V.S.F. Garnot J. Björklund L. Schneider J.D. Wegner G. von Arx Towards ROXAS AI: deep learning for faster and more accurate conifer cell analysis Dendrochronologia 81 126126 10.1016/j.dendro.2023.126126
A. Garcia-Pedrero A.I. García-Cervigón J.M. Olano M. García-Hidalgo M. Lillo-Saavedra C. Gonzalo-Martín et al. Convolutional neural networks for segmenting xylem vessels in stained cross-sectional images Neural Comput Appl 32 24 17927 39 10.1007/s00521-019-04546-6
L. Verschuren T. De Mil P. De Frenne K. Haneca J. Van Acker K. Vandekerkhove et al. Heading for a fall: the fate of old wind-thrown beech trees (Fagus sylvatica) is detectable in their growth pattern Sci Total Environ 903 166148 1:CAS:528:DC%2BB3sXhslSmtb%2FL 10.1016/j.scitotenv.2023.166148 37574075
Dwyer B, Nelson J, Hansen T, et al. Roboflow. https://roboflow.com.
G. Van Rossum F.L. Drake Python 3 reference manual Scotts Valley, CA CreateSpace
M. De Ridder J. Van den Bulcke D. Vansteenkiste D. Van Loo M. Dierick B. Masschaele et al. High-resolution proxies for wood density variations in Terminalia superba Ann Bot 107 2 293 302 10.1093/aob/mcq224 21131386
T. De Mil A. Vannoppen H. Beeckman J. Van Acker J. Van den Bulcke A field-to-desktop toolchain for X-ray CT densitometry enables tree ring analysis Ann Bot 117 mcw063 10.1093/aob/mcw063
J. Van den Bulcke E.L.G. Wernersson M. Dierick D. Van Loo B. Masschaele L. Brabant et al. 3D tree-ring analysis using helical X-ray tomography Dendrochronologia 32 1 39 46 10.1016/j.dendro.2013.07.001
J. Schindelin I. Arganda-Carreras E. Frise V. Kaynig M. Longair T. Pietzsch et al. FIJI: an open-source platform for biological-image analysis Nat Methods 9 7 676 82 1:CAS:528:DC%2BC38XhtVKnurbJ 10.1038/nmeth.2019 22743772
A. Rita J.J. Camarero M. Colangelo E.G. de Andrés M. Pompa-García Wood anatomical traits respond to climate but more individualistically as compared to radial growth: analyze trees, not means Forests 13 6 956 10.3390/f13060956
M. Carrer Individualistic and time-varying tree-ring growth to climate sensitivity PLoS One 6 7 e22813 1:CAS:528:DC%2BC3MXhtVyks7bK 10.1371/journal.pone.0022813 21829523 3145760
P. Fonti I. García-González Suitability of chestnut earlywood vessel chronologies for ecological studies New Phytol 163 1 77 86 10.1111/j.1469-8137.2004.01089.x 33873786
H. Gärtner D. Nievergelt The core-microtome: a new tool for surface preparation on cores and time series analysis of varying cell parameters Dendrochronologia 28 2 85 92 10.1016/j.dendro.2009.09.002
J. Chalfoun M. Majurski T. Blattner K. Bhadriraju W. Keyrouz P. Bajcsy et al. MIST: accurate and scalable microscopy image stitching tool with stage modeling and error minimization Sci Report 7 1 4988 10.1038/s41598-017-04567-y
S. Preibisch S. Saalfeld P. Tomancak Globally optimal stitching of tiled 3D microscopic image acquisitions Bioinformatics 25 11 1463 5 1:CAS:528:DC%2BD1MXmtFeqtrw%3D 10.1093/bioinformatics/btp184 19346324 2682522
R. Guay WinCELL 2013 for wood cell analysis Ontario Regent Instruments Canada Inc
G. Von Arx M. Carrer ROXAS-A new tool to build centuries-long tracheid-lumen chronologies in conifers Dendrochronologia 32 3 290 3 10.1016/j.dendro.2013.12.001
K. Wiśniewská K. Šilhán Quantitative anatomy or macroscopic parameters of compression wood of Picea abies (L.) H. Karst.? Defining the optimal parameters for dendrogeomorphic purposes IAWA Journal 44 1 21 35 10.1163/22941932-bja10098
D. Castagneri L. Regev E. Boaretto M. Carrer Xylem anatomical traits reveal different strategies of two Mediterranean oaks to cope with drought and warming Environ Exp Bot 133 128 38 10.1016/j.envexpbot.2016.10.009
I. Arganda-Carreras V. Kaynig C. Rueden K.W. Eliceiri J. Schindelin A. Cardona et al. Trainable Weka Segmentation: a machine learning tool for microscopy pixel classification Bioinformatics 33 15 2424 6 1:CAS:528:DC%2BC1cXitFOntb%2FO 10.1093/bioinformatics/btx180 28369169
T. De Mil Y. Tarelkin S. Hahn W. Hubau V. Deklerck O. Debeir et al. Wood density profiles and their corresponding tissue fractions in tropical angiosperm trees Forests 9 12 763 10.3390/f9120763
P. Bankhead M.B. Loughrey J.A. Fernández Y. Dombrowski D.G. McArt P.D. Dunne et al. QuPath: open source software for digital pathology image analysis Sci Reports 7 1 1 7 1:CAS:528:DC%2BC1cXhsFamsL%2FE
R. Keret P.M. Schliephack D.F. Stangler T. Seifert H.P. Kahle D.M. Drew et al. An open-source machine-learning approach for obtaining high-quality quantitative wood anatomy data from E. grandis and P. radiata xylem Plant Sci 340 111970 1:CAS:528:DC%2BB2cXhtF2htL8%3D 10.1016/j.plantsci.2023.111970 38163623
K. Ziemińska D.W. Butler S.M. Gleason I.J. Wright M. Westoby Fibre wall and lumen fractions drive wood density variation across 24 Australian angiosperms AoB Plants 5 PLT046 10.1093/aobpla/plt046 4104653
B. Forster D. Van De Ville J. Berent D. Sage M. Unser Complex wavelets for extended depth-of-field: a new method for the fusion of multichannel microscopy images Microscopy Res Tech 65 1–2 33 42 10.1002/jemt.20092
T. Peng K. Thorn T. Schroeder L. Wang F.J. Theis C. Marr et al. A BaSiC tool for background and shading correction of optical microscopy images Nat Commun 8 1 14836 1:CAS:528:DC%2BC2sXpvFemtbc%3D 10.1038/ncomms14836 28594001 5472168
