基于网络图的地基激光雷达复杂树木点云枝叶分离方法

林筱涵; 李爱农; 边金虎; 张正建; 南希

doi:10.11873/j.issn.1004-0323.2022.1.0161

閬ユ劅鎶�鏈笌搴旂敤鈥衡�� 2022, Vol. 37 鈥衡�� Issue (1): 161-172.DOI: 10.11873/j.issn.1004-0323.2022.1.0161

鈥� 闈掍績浼氬崄鍛ㄥ勾涓撴爮鈥� 涓婁竴绡� 涓嬩竴绡�

鍩轰簬缃戠粶鍥剧殑鍦板熀婵�鍏夐浄杈惧鏉傛爲鏈ㄧ偣浜戞灊鍙跺垎绂绘柟娉�

鏋楃娑�^1,²(),鏉庣埍鍐�¹(),杈归噾铏�¹,寮犳寤�^1,²,鍗楀笇¹

^1.涓浗绉戝闄㈡按鍒╅儴鎴愰兘灞卞湴鐏惧涓庣幆澧冪爺绌舵墍鏁板瓧灞卞湴涓庨仴鎰熷簲鐢ㄤ腑蹇冿紝鍥涘窛鎴愰兘 610041
^2.涓浗绉戝闄㈠ぇ瀛﹁祫婧愪笌鐜瀛﹂櫌锛屽寳浜� 100049

鏀剁鏃ユ湡:2021-10-08 淇洖鏃ユ湡:2022-01-05 鍑虹増鏃ユ湡:2022-02-20 鍙戝竷鏃ユ湡:2022-04-08
閫氳浣滆��: 鏉庣埍鍐�
浣滆�呯畝浠�:鏋楃娑碉紙1996鈥旓級锛屽北涓滃▉娴蜂汉锛岀敺锛岀澹爺绌剁敓锛屼富瑕佷粠浜嬫縺鍏夐浄杈鹃仴鎰熺爺绌躲�侲?mail锛�linxiaohan@imde.ac.cn
鍩洪噾璧勫姪:
鍥藉閲嶇偣鐮斿彂璁″垝椤圭洰鈥滃北鍦扮敓鎬佺郴缁熷叏鐞冨彉鍖栧叧閿弬鏁扮珛浣撹娴嬩笌楂樺垎杈ㄧ巼浜у搧鐮斿埗鈥�(2020YFA0608700);鍥藉鑷劧绉戝鍩洪噾椤圭洰鈥滃北鍦板吀鍨嬬敓鎬佸弬閲忛仴鎰熷弽婕斿缓妯″強鍏舵椂绌鸿〃寰佽兘鍔涚爺绌垛��(41631180)

A Method for Separating Leaf and Wood Components of Complex Tree Point Cloud Data based on Network Graph with Terrestrial Laser Scanning

Xiaohan Lin^1,²(),Ainong Li¹(),Jinhu Bian¹,Zhengjian Zhang^1,²,Xi Nan¹

^1.Center for Digital Mountain and Remote Sensing Application锛孖nstitute of Mountain Hazards and Environment锛孋hinese Academy of Sciences锛孋hengdu 610041锛孋hina
^2.University of Chinese Academy of Sciences锛孊eijing 100049锛孋hina

Received:2021-10-08 Revised:2022-01-05 Online:2022-02-20 Published:2022-04-08
Contact: Ainong Li

鎽樿/Abstract

鎽樿锛�

鍦板熀婵�鍏夐浄杈炬爲鏈ㄧ偣浜戞暟鎹殑鏋濆彾鍒嗙鏄簿纭绠楀湴涓婄敓鐗╅噺鍜屽彾闈㈢Н鎸囨暟鐨勯噸瑕佸墠鎻愶紝涔熸槸鏍戞湪涓夌淮寤烘ā鐨勯噸瑕佹楠ゃ�傜劧鑰岋紝灞卞湴澶嶆潅鏍戞湪鍐犲箙澶т笖缁撴瀯澶嶆潅锛屼粠鑰岄�犳垚鏍戝彾涓庢灊骞蹭箣闂寸殑鐩镐簰閬尅锛屽洜姝ゅ緢闅捐幏鍙栭珮璐ㄩ噺鐨勭偣浜戞暟鎹紝鐩墠瀵瑰叾瀹炵幇鏋濆彾鍒嗙渚濈劧瀛樺湪杈冨ぇ鐨勫洶闅俱�傚埄鐢ㄥ湴鍩烘縺鍏夐浄杈綟ARO Focus^3D X330鑾峰彇涓夌淮婵�鍏夌偣浜戞暟鎹紝鎻愬嚭浜嗕竴绉嶅熀浜庣綉缁滃浘鐨勬爲鏈ㄧ偣浜戞灊鍙跺垎绂绘柟娉曘�傞鍏堬紝閲囩敤LeWos妯″瀷瀵圭偣浜戣繘琛屽垵姝ョ殑鏋濆彾鍒嗙锛屽垎绂诲嚭鏋濆共鍜屽彾鐗囩偣浜戙�傚湪姝ゅ熀纭�涓婏紝閽堝鏋濆共鍜屽彾鐗囨贩鍚堢偣浜戦�氳繃璺緞杩借釜妫�娴嬬畻娉曟潵绮剧粏鍒嗙鏋濆共鍜屽彾鐗囥�傞殢鐫�璺緞闀垮害浠�10澧炲姞鍒�100锛屾灊骞茬偣涓嶆柇澧炲姞锛屽彾鐗囩偣涓嶆柇鍑忓皯锛屾灊鍙跺垎绂荤簿纭害銆佹灊骞睩鍒嗘暟銆佸彾鐗嘑鍒嗘暟銆並appa绯绘暟鍧囧厛澧炲姞鍚庡噺灏戙�傜患鍚堣繖4椤圭簿搴﹁瘎浠锋寚鏍囷紝閫夊彇鍚勪釜鏍戞湪鏈�浼樿矾寰勯暱搴︽墽琛岃矾寰勮拷韪娴嬬畻娉曘�傞�氳繃涓嶭eWos妯″瀷銆乀lseparation妯″瀷鍜岄珮鏂贩鍚堟ā鍨嬬瓑涓绘祦鏋濆彾鍒嗙鏂规硶姣旇緝鍙戠幇锛岃鏂规硶绮惧害鏇翠紭锛岀簿纭害涓�91.97%銆傝�屼笖锛岃鏂规硶鐨勬灊骞睩鍒嗘暟鍜屽彾鐗嘑鍒嗘暟鍧囧ぇ浜�85%锛岃繖琛ㄦ槑璇ユ柟娉曞叿鏈夊緢濂界殑骞宠　鎬с�傝鏂规硶浠呬娇鐢ㄨ矾寰勯暱搴︼紝涓嶈�冭檻鍑犱綍鐗瑰緛锛屽洜姝ゆ瀬澶у湴鎻愰珮浜嗛拡鍙舵爲鏈ㄧ殑鏋濆彾鍒嗙绮惧害銆傚悓鏃讹紝璇ユ柟娉曞涓嶅悓鏍戠銆佷笉鍚屽瘑搴︾偣浜戞灊鍙跺垎绂绘晥鏋滃潎杈冨ソ锛岄瞾妫掓�у己銆傜簿纭殑鏍戞湪鐐逛簯鏋濆彾鍒嗙瀵规．鏋楄祫婧愮鐞嗗拰鐢熺墿澶氭牱鎬х爺绌剁瓑鍏锋湁閲嶈鎰忎箟銆�

鍏抽敭璇�: 鍦板熀婵�鍏夐浄杈�, 鏋濆彾鍒嗙, 缃戠粶鍥�, LeWos妯″瀷, 璺緞杩借釜妫�娴嬬畻娉�

Abstract:

The leaf and wood separation of the terrestrial laser scanning tree point cloud data is an important prerequisite for the accurate estimation of above-ground biomass and leaf area index锛� and it is also an important step for three-dimensional modeling of a tree. However锛� complex trees in mountain areas have large crowns and complex structures锛� resulting in mutual occlusion between leaves and branches. Therefore锛� it is difficult to obtain high-quality point cloud data. At present锛� it is still difficult for complex trees to separate leaf and wood components. High-resolution point clouds were acquired with Faro Focus^3D X330. This paper proposes a method for leaf and wood separation of tree point cloud based on network graph. First锛� the LeWos model is used to perform preliminary leaf and wood separation on the point cloud锛� and separate the wood and leaf point cloud. On this basis锛� the path retrace detection algorithm is used to finely separate the leaf and wood for the mixed point clouds. As the retrace steps increases from 10 to 100锛� the wood points continue to increase锛� the leaf points continue to decrease锛� the accuracy锛� wood F-score decreases锛� leaf F-score and the Kappa coefficient first increases and then decreases. By comparing with LeWos model锛� Tlseparation model and Gaussian mixture model锛� it is found that the research method in this paper has the better precision锛� with an accuracy of 91.97%. Moreover锛� the wood F-score and leaf F-score of the proposal method are both greater than 85%锛� which means that the proposal method has a good balance when classifying wood and leaf. The proposal method only uses the retrace steps and does not consider the geometric characteristics锛� so the accuracy of branch and leaf separation of coniferous trees is greatly improved. At the same time锛� the method in this paper has a good effect on the separation of wood and leaf components of point clouds with different densities and different species. Therefore锛� the method in this paper is more robust. Accurate wood and leaf separation of tree point clouds is of great significance to forest resource management and biodiversity research.

Key words: Terrestrial laser scanning, Leaf and wood separation, Network graph, LeWos model, Path trace detection algorithm

涓浘鍒嗙被鍙�:

TN958.98

鏋楃娑�,鏉庣埍鍐�,杈归噾铏�,寮犳寤�,鍗楀笇. 鍩轰簬缃戠粶鍥剧殑鍦板熀婵�鍏夐浄杈惧鏉傛爲鏈ㄧ偣浜戞灊鍙跺垎绂绘柟娉昜J]. 閬ユ劅鎶�鏈笌搴旂敤, 2022, 37(1): 161-172.

Xiaohan Lin,Ainong Li,Jinhu Bian,Zhengjian Zhang,Xi Nan. A Method for Separating Leaf and Wood Components of Complex Tree Point Cloud Data based on Network Graph with Terrestrial Laser Scanning[J]. Remote Sensing Technology and Application, 2022, 37(1): 161-172.

鍙傝�冩枃鐚�

1	Dassort M锛� Constant T锛� Fournier M. The use of terrestrial LiDAR technology in forest science锛� application fields锛� benefits and challenges锛籎锛�. Annal of Forest Science锛� 2011锛� 68锛�5锛夛細959-974. DOI锛� 10.1007/s13595-011-0102-2 .
2	Ma Y C锛� Piao S L锛� Sun Z Z锛� et al. Stand ages regulate the response of soil respiration to temperature in a Larix Principis-rupprechtii plantation锛籎锛�. Agricultural and Forest Meteorology锛�2014锛�184锛�179-187. DOI锛�10.1016/j.agrformet.2013. 10.008 .
3	Srinivasan S锛� Popescu S C锛� Eriksson M锛� et al. Multi-temporal terrestrial laser scanning for modeling tree biomass change锛籎锛�. Foestr Ecology and Management锛� 2014锛� 318锛� 304-317. DOI锛� 10.1016/j.foreco.2014.01.038 .
4	Liang X L锛� Kankare V锛� Hyyppa J锛� et al. Terrestrial laser scanning in forest inventories锛籎锛�. ISPRS Journal of Photogrammetry and Remote Sensing锛� 2016锛� 115锛�63-77. DOI锛� 10.1016/j.isprsjprs.2016.01.006 .
5	Lin Y. LiDAR锛� An important tool for next-generation phenotyping technology of high potential for plant phenomics锛燂蓟J锛�. Computers and Electronics and Agriculture锛� 2015锛� 119锛�61-73.DOI锛� 10.1016/j.compag.2015.10.011 .
6	Llorens J锛� Gil E锛� Llop J锛� et al. Ultrasonic and LiDAR sensors for electronic canopy characterization in vineyards锛� advances to improve pesticide application methods锛籎锛�. Sensors锛� 2011锛� 11锛�2锛夛細 2177-2194.DOI锛� 10.3390/s110202177 .
7	Msdec S锛� Baret F锛� de Solan B锛� et al. High-throughput phenotyping of plant height锛� comparing unmanned aerial vehicles and ground LiDAR estimates锛籎锛�. Frontiers in Plant Science锛� 2017锛� 8锛�14.DOI锛� 10.3389/fpls.2017.02002 .
8	Berk P锛� Hocevar M锛� Stajnko D锛� et al. Development of alternative plant protection product application techniques in Orchards锛� based on measurement sensing systems锛� a review锛籎锛�. Computers and Electronics in Agriculture锛� 2016锛� 124锛� 273-288.DOI锛� 10.1016/j.compag.2016.04.018 .
9	Yu Y锛� Gao T锛� Zhu J锛� et al. Terrestrial laser scanning鈥恉erived Canopy interception index for predicting rainfall interception锛籎锛�. Ecohydrology锛�2020锛�13锛�5锛夛細15. DOI锛�10.1002/eco. 2212 .
10	Li Y M锛� Su Y J锛� ZHao X X锛� et al. Retrieval of tree branch architecture attributes from terrestrial laser scan data using a Laplacian Algorithm锛籎锛�. Agricultural and Forest Meteorology锛� 2020锛� 284锛�107874.DOI锛� 10.1016/j.agrformet.2019.107874 .
11	Li Y M锛� Guo Q H锛� Su Y J锛� et al. Retrieving the gap fraction锛� element clumping index锛� and leaf area index of individual trees using single-scan data from a terrestrial laser scanner锛籎锛�. ISPRS Journal of Photogrammetry and Remote Sensing锛� 2017锛� 130锛�308-316.DOI锛� 10.1016/j.isprsjprs.2017.06.006 .
12	Zhao K锛� Garcia M锛� Liu S锛� et al. Terrestrial Lidar remote sensing of forests锛� maximum likelihood estimates of canopy profile锛� leaf area index锛� and leaf angle distribution锛籎锛�. Agricultural and Forest Meteorology锛�2015锛�209-210锛�100-113. DOI锛� 10.1016/j.agrformet.2015.03.008 .
13	Bailey B N锛� Mahaffee W F. Rapid measurement of the three-dimensional distribution of leaf orientation and the leaf angle probability density function using terrestrial LiDAR scanning锛籎锛�.Remote Sensing of Environment锛�2017锛�194锛�63-76. DOI锛� 10.1088/1361-6501/aa5cfd .
14	Liu J锛� Wang T锛� Skidmore A K锛� et al. Comparison of terrestrial LiDAR and digital hemispherical photography for estimating leaf angle distribution in european broadleaf beech forests锛籎锛�. ISPRS Journal of Photogrammetry and Remote Sensing锛� 2019锛� 158锛�76-89. DOI锛� 10.1016/j.isprsjprs.2019.09.015 .
15	C?t茅 J F锛� Widlowski J L锛� Fournier R A锛� et al. The structural and radiative consistency of three-dimensional tree reconstructions from terrestrial LiDAR锛籎锛�. Remote Sensing of Environment锛�2009锛�113锛�5锛夛細1067-1081. DOI锛�10.1016/j.rse.2009. 01.017 .
16	Hu S J锛� Li Z R锛� Zhang Z Y锛� et al. Efficient tree modeling from airborne LiDAR point clouds锛籎锛�. Computers & Graphics锛� 2017锛� 67锛�1-13.DOI锛� 10.1016/j.cag.2017.04.004 .
17	Calders K锛� Origo N锛� Burt A锛� et al. Realistic forest stand reconstruction from terrestrial LiDAR for radiative transfer modelling锛籎锛�. Remote Sensing锛� 2018锛� 10锛�6锛夛細 15.DOI锛� 10.3390/rs10060933 .
18	Tao S L锛� Guo Q H锛� Xu S W锛� et al. A geometric method for wood-leaf separation using terrestrial and simulated LiDAR data锛籎锛�. Photogrammetric Engineering & Remote Sensing锛� 2015锛� 81锛�10锛夛細 767-776.DOI锛� 10.14358/pers.81.10.767 .
19	Strahler A H锛� Jupp D L B锛� Woodcock C E锛� et al. Retrieval of forest structural parameters using a ground-based LiDAR instrument 锛圗chidna 锛圧锛夛級锛籎锛�. Canadian Journal of Remote Sensing锛� 2008锛� 34锛歋426-S40.DOI锛� 10.5589/m08-046 .
20	Danson F M锛� Sasse F锛� Schofield L A. Spectral and spatial information from a novel dual-wavelength full-waveform terrestrial laser scanner for forest ecology锛籎锛�.Interface Focus锛�2018锛� 8锛�2锛夛細 20170049.DOI锛� 10.1098/rsfs.2017.0049 .
21	Raumonen P锛� Kaasalainen M锛� Akerblom M锛� et al. Fast automatic precision tree models from terrestrial laser scanner data锛籎锛�. Remote Sensing锛� 2013锛� 5锛�2锛夛細 491-520.DOI锛� 10.3390/rs5020491 .
22	Calders K锛� Disney M I锛� Armston J锛� et al. Evaluation of the range accuracy and the radiometric calibration of multiple terrestrial laser scanning instruments for data interoperability锛籎锛�. IEEE Transactions on Geoscience and Remote Sensing锛� 2017锛�55锛�5锛夛細2716-2724. DOI锛�10.1109/tgrs.2017.2652721 .
23	Disney M I锛� Vicari M B锛� Burt A锛� et al. Weighing trees with lasers锛� advances锛� challenges and opportunities锛籎锛�. Interface Focus锛� 2018锛� 8锛�2锛夛細 14.DOI锛� 10.1098/rsfs.2017.0048 .
24	Yun T锛� An F锛� Li W锛� et al. A novel approach for retrieving tree leaf area from ground-based LiDAR锛籎锛�. Remote Sensing锛� 2016锛� 8锛�11锛夛細 21.DOI锛� 10.3390/rs8110942 .
25	Ma L X锛� Zheng G锛� Eitel J U H锛� et al. Improved salient feature-based approach for automatically separating photosynthetic and nonphotosynthetic components within terrestrial LiDAR point cloud data of forest canopies锛籎锛�. IEEE Transactions on Geoscience and Remote Sensing锛� 2016锛� 54锛�2锛夛細 679-696.DOI锛� 10.1109/tgrs.2015.2459716 .
26	Wang D锛� Hollaus M锛� Pfeifer N. Feasibility of machine learning methods for separating wood and leaf points from terrestrial laser scaning data锛籎锛�. ISPRS Annals of Photogrammetry锛� Reomte Sensing and Spatial Information Sciences锛�2017锛� IV-2/W4锛�157-164.
27	Li S H锛� Dai L Y锛� Wang H S锛� et al. Estimating leaf area density of individual trees using the point cloud segmentation of terrestrial LiDAR data and a Voxel-based model锛籎锛�. Remote Sensing锛� 2017锛� 9锛�11锛夛細 1202.DOI锛� 10.3390/rs9111202 ..
28	Vicari M B锛� Disney M锛� Wilkes P锛� et al. Leaf and wood classification framework for terrestrial LiDAR point clouds锛籎锛�. Methods in Ecology and Evolution锛� 2019锛� 10锛�5锛夛細 680-694.DOI锛� 10.1111/2041-210x.13144 .
29	Su Z H锛� Li S H锛� Liu H H锛� et al. Extracting wood point cloud of individual trees based on geometric features锛籎锛�. IEEE Geoscience and Remote Sensing Letters锛� 2019锛� 16锛�8锛夛細 1294-1298.DOI锛� 10.1109/lgrs.2019.2896613 .
30	Wang Hongshu锛� Li Shihua锛� Guo Jiawei锛� et al. Retrieval of the leaf area density of magnolia woody canopy with terrestrial laser-scanning data锛籎锛�.Journal of Remote Sensing锛�2016锛�20锛�4锛夛細570-578.
30	鐜嬫椽铚�锛屾潕涓栧崕锛岄儹鍔犱紵锛岀瓑. 鍦板熀婵�鍏夐浄杈剧殑鐜夊叞鏋楀啝鍙堕潰绉瘑搴﹀弽婕旓蓟J锛�.閬ユ劅瀛︽姤锛�2016锛�20锛�4锛夛細570-578
31	Wang D锛� Takoudjou S M锛� Casella E. LeWoS锛� A universal leaf-wood classification method to facilitate the 3D modelling of large tropical trees using terrestrial LiDAR锛籎锛�. Methods in Ecology and Evolution锛�2020锛�11锛�3锛夛細376-389. DOI锛�10.1111/2041-210x.13342 .
32	Guinard S锛� Landrieu L. Weakly supervised segementation-alded classification of urban scences from 3D LIDAR point clouds锛籑锛�. Gottingen. Copernicus Gesellschaft Mbh. 2017.
33	Wang D锛� Brunner J锛� Ma Z Y锛� et al. Separating tree photosynthetic and non-photosynthetic components from point cloud data using dynamic segment merging锛籎锛�. Forests锛�2018锛�9锛�5锛夛細23. DOI锛� 10.3390/f9050252 .
34	Klasing K锛� Wollherr D锛� Buss M锛� et al. A clustering method for efficient segmentation of 3 D laser data锛籑锛�. New York锛� IEEE. 2008.
35	Raumonen P锛� Tarvainen T. Segmentation of vessel structures from photoacoustic images with reliability assessment锛籎锛�. Biomedical Optics Express锛�2018锛�9锛�7锛夛細2887-2904. DOI锛� 10. 1364/boe.9.002887 .
36	Landrieu L锛� Raguet H锛� Vallet B锛� et al. A structured regularization framework for spatially smoothing semantic labelings of 3D point clouds锛籎锛�. ISPRS Journal of Photogrammetry and Remote Sensing锛�2017锛�132锛�102-118. DOI锛�10.1016/j.isprsjprs. 2017.08.010 .
37	Wang Z锛� Zhang L Q锛� Fang T锛� et al. A multiscale and hierarchical feature extraction method for terrestrial laser scanning point cloud classification锛籎锛�. IEEE Transactions on Geoscience and Remote Sensing锛�2015锛�53锛�5锛夛細2409-2425. DOI锛�10. 1109/tgrs.2014.2359951 .
38	Sokolova M锛� Japkowicz N锛� Szpakowicz S. Beyond accuracy锛� F-Score and ROC锛� a family of discriminant measures for performance evaluation锛籆锛解垾 Australian Joint Conference on Artificial Intelligence锛� Advances in Artificial Intelligence锛孒obart Australia锛� 2006.
39	de Tanago J G锛� Lau A锛� Bartholomeus H锛� et al. Estimation of above-ground biomass of large tropical trees with terrestrial LiDAR锛籎锛�. Methods in Ecology and Evolution锛� 2018锛� 9锛�2锛夛細 223-234. DOI锛� 10.1111/2041-210x.12904 .
40	Paynter I锛� Genest D锛� Saenz E锛� et al. Quality assessment of terrestrial laser scanner ecosystem observations using pulse trajectories锛籎锛�. IEEE Transactions on Geoscience and Remote Sensing锛� 2018锛� 56锛�11锛夛細 6324-6333. DOI锛� 10.1109/tgrs.2018.2836947 .
41	Wan P锛� Zhang W M锛� Jin S N锛� et al. Plot-level wood-leaf separation of trees using terrestrial LiDAR data based on a segmentwise geometric feature classification method锛籎锛�. 2020. DOI锛� 10.21203/rs.3.rs-42032/v1 .

[1]	寮犲缓楣�,鐜嬮噾浜�,鍒樺箍鏉�,楹诲崼宄�,鍒橀挶濞�,閭撲簯绋�. 鍩轰簬鍦板熀闆疯揪鐐逛簯涓绘柟鍚戠殑鏋椾笅妞嶈鑷姩婊ら櫎[J]. 閬ユ劅鎶�鏈笌搴旂敤, 2023, 38(2): 405-412.
[2]	宕斿槈鏂�,闄堝仴,鎴愰珮娣�,鐭虫弧,鍒樹匠鐞�,绁濆杽鍙�. 鍗忓悓鍦板熀鍜岀┖鍩烘縺鍏夐浄杈剧殑娌欏皹澶╂皵瑙傛祴鐮旂┒[J]. 閬ユ劅鎶�鏈笌搴旂敤, 2022, 37(2): 416-423.

鍩轰簬缃戠粶鍥剧殑鍦板熀婵�鍏夐浄杈惧鏉傛爲鏈ㄧ偣浜戞灊鍙跺垎绂绘柟娉�

A Method for Separating Leaf and Wood Components of Complex Tree Point Cloud Data based on Network Graph with Terrestrial Laser Scanning

RichHTML

PDF (PC)

鍙鍖�

鎽樿/Abstract

寮曠敤鏈枃

浣跨敤鏈枃

鍙傝�冩枃鐚�

鐩稿叧鏂囩珷 2

缂栬緫鎺ㄨ崘

Metrics

鏈枃璇勪环