南京林业大学学报(自然科学版) ›› 2024, Vol. 48 ›› Issue (4): 104-112.doi: 10.12302/j.issn.1000-2006.202205041
所属专题: 专题报道Ⅲ:智慧林业之森林可视化研究
• 专题报道Ⅲ:智慧林业之森林可视化研究(执行主编 李凤日、张怀清、曹林) • 上一篇 下一篇
收稿日期:
2022-05-26
修回日期:
2022-10-05
出版日期:
2024-07-30
发布日期:
2024-08-05
通讯作者:
*林文树(linwenshu@nefu.edu.cn),教授。作者简介:
钟浩(260919837@qq.com),博士生。
基金资助:
ZHONG Hao(), WANG Chuhong, LIN Wenshu*()
Received:
2022-05-26
Revised:
2022-10-05
Online:
2024-07-30
Published:
2024-08-05
摘要:
【目的】应用遥感技术进行树种识别已成为森林调查的重要手段之一,但利用近地面遥感手段进行树种识别时存在地基激光雷达(LiDAR)点云数据树冠上层信息不足和无人机影像树冠下层的信息缺失的问题,联合多源遥感数据快速准确识别树种,对研究和保护森林资源具有重要意义。【方法】以哈尔滨市东北林业大学城市林业示范基地中樟子松(Pinus sylvestris var. mongolica)、黑皮油松(P. tabuliformis var. mukdensis)、水曲柳(Fraxinus mandshurica)、胡桃楸(Juglans mandshurica)为研究对象进行树种识别研究,获取地基LiDAR与无人机影像数据,通过对高重叠度无人机影像进行处理得到无人机影像点云,将无人机影像点云与地基LiDAR点云数据进行配准融合,对融合后数据进行单木分割得到单木点云,基于单木点云提取形状特征、结构特征、树干颜色特征和树冠颜色特征,借助支持向量机分类算法进行树种识别并结合随机森林算法对不同特征的识别能力进行了分析。【结果】利用所有特征进行树种识别取得试验最优结果,其总精度和Kappa 系数分别为93.48%和0.91。相较于其他对比方案,其总精度和Kappa 系数分别提升4.35~16.31个百分点和0.06~0.22。【结论】提出了一种地基LiDAR数据与无人机影像点云数据进行融合的树种识别方法,该方法能够在一定程度上弥补树种识别中特征提取时地基LiDAR点云数据树冠上层信息不足,以及无人机影像树冠下层信息缺失等问题。充分利用多源数据所包含的丰富信息进行树种识别,可有效提高树种识别精度。
中图分类号:
钟浩,王楚虹,林文树. 联合地基激光雷达与无人机影像的树种识别[J]. 南京林业大学学报(自然科学版), 2024, 48(4): 104-112.
ZHONG Hao, WANG Chuhong, LIN Wenshu. Tree species identification of combined TLS date and UAV images[J].Journal of Nanjing Forestry University (Natural Science Edition), 2024, 48(4): 104-112.DOI: 10.12302/j.issn.1000-2006.202205041.
表2
融合前后树木参数提取结果"
树种 tree species | 融合类型 fusion type | DBH/ cm | HT/ m | LC/ m | AC/ m2 | VC/m3 |
---|---|---|---|---|---|---|
樟子松 P. sylvestnis var. mongolica | 融合 | 26.1 | 16.68 | 4.03 | 13.28 | 59.44 |
未融合 | 15.28 | 3.65 | 11.32 | 42.58 | ||
黑皮油松 P. tabuliformis var. mukdensis | 融合 | 23.6 | 13.52 | 5.26 | 23.58 | 103.35 |
未融合 | 13.01 | 4.90 | 20.79 | 79.77 | ||
水曲柳 F. mandshurica | 融合 | 29.5 | 20.71 | 6.35 | 33.93 | 172.48 |
未融合 | 20.13 | 6.06 | 31.01 | 147.01 | ||
胡桃楸 J. mandshurica | 融合 | 33.4 | 17.37 | 7.16 | 42.96 | 206.27 |
未融合 | 16.37 | 6.64 | 37.46 | 174.57 |
表3
树种识别结果"
方案 scheme | 参数 parameter | 各树种识别精度/% accuracy | 总精度/% OA | Kappa系数 Kappa coefficient | |||
---|---|---|---|---|---|---|---|
樟子松 P. sylvestris var. mogolica | 黑皮油松 P. tabuliformis var. mukdensis | 水曲柳 F. mandshurica | 胡桃楸 J. mandshurica | ||||
1 | 生产者精度PA | 83.33 | 50.00 | 73.91 | 92.59 | 77.17 | 0.69 |
用户精度UA | 71.43 | 69.23 | 73.91 | 89.29 | |||
2 | 生产者精度PA | 91.67 | 83.33 | 78.26 | 77.78 | 82.61 | 0.77 |
用户精度UA | 81.48 | 78.95 | 81.82 | 87.50 | |||
3 | 生产者精度PA | 91.67 | 72.22 | 82.61 | 92.59 | 85.87 | 0.81 |
用户精度UA | 91.67 | 76.47 | 90.48 | 83.33 | |||
4 | 生产者精度PA | 95.83 | 88.89 | 82.61 | 85.19 | 88.04 | 0.84 |
用户精度UA | 88.46 | 94.12 | 82.61 | 88.46 | |||
5 | 生产者精度PA | 83.33 | 55.56 | 86.96 | 92.59 | 81.52 | 0.75 |
用户精度UA | 83.33 | 71.43 | 76.92 | 89.29 | |||
6 | 生产者精度PA | 83.33 | 94.44 | 86.96 | 92.59 | 89.13 | 0.85 |
用户精度UA | 83.33 | 85.00 | 90.91 | 96.15 | |||
7 | 生产者精度PA | 95.83 | 94.44 | 86.96 | 96.30 | 93.48 | 0.91 |
用户精度UA | 92.00 | 94.44 | 95.24 | 92.86 |
图5
特征重要性排序及树种识别精度 C-Red-Mean.树冠红色波段平均值 canopy red band mean;C-Blue-Mean.树冠蓝色波段平均值 canopy blue band mean;T-Red-Mean.树干红色波段平均值 trunk red band mean;T-Blue-Mean.树干蓝色波段平均值 trunk blue band mean;T-Green-Mean.树干绿色波段平均值 trunk green band mean;C-Red-Std.树冠红色波段标准差 canopy red band standard deviation;C-Green-Mean.树冠绿色波段平均值 canopy green band mean;T-Red-Std.树干红色波段标准差 trunk red band standard deviation;RL/H.冠幅树高比 canopy width to height ratio;C-Blue-Std.树冠蓝色波段标准差 canopy blue band standard deviation;T-Green-Std.树干绿色波段标准差 trunk green band standard deviation;T-Blue-Std.树干蓝色波段标准差 trunk blue band standard deviation;RA/H.树冠面积树高比 canopy area to height ratio;C-Green-Std.树冠绿色波段标准差 canopy green band standard deviation;RA/D.树冠面积胸径比 canopy area to diameter at breast height ratio;RV/D.树冠体积胸径比 canopy volume to diameter at breast height ratio;RV/H.树冠体积树高比 canopy volume to height ratio;RL/D.冠幅胸径比 canopy width to diameter at breast height ratio;0~10%、10%~20%、…、90%~100%表示分位点云数量占比0~10%,10%~20%,…,90%~100% represent quantile point cloud ratio。"
[1] | LATIFI H, HEURICH M. Multi-scale remote sensing-assisted forest inventory:a glimpse of the state-of-the-art and future prospects[J]. Remote Sens, 2019, 11(11):1260.DOI: 10.3390/rs11111260. |
[2] | WHITE J C, COOPS N C, WULDER M A, et al. Remote sensing technologies for enhancing forest inventories:a review[J]. Can J Remote Sens, 2016, 42(5):619-641.DOI: 10.1080/07038992.2016.1207484. |
[3] | 李增元, 陈尔学. 中国林业遥感发展历程[J]. 遥感学报, 2021, 25(1):292-301. |
LI Z Y, CHEN E X. Development course of forestry remote sensing in China[J]. Natl Remote Sens Bull, 2021, 25(1):292-301. | |
[4] | FENG B K, ZHENG C, ZHANG W Q, et al. Analyzing the role of spatial features when cooperating hyperspectral and LiDAR data for the tree species classification in a subtropical plantation forest area[J]. JARS, 2020, 14(2):022213.DOI: 10.1117/1.JRS.14.022213. |
[5] | 曹林, 佘光辉, 代劲松, 等. 激光雷达技术估测森林生物量的研究现状及展望[J]. 南京林业大学学报(自然科学版), 2013, 37(3):163-169. |
CAO L, SHE G H, DAI J S, et al. Status and prospects of the LiDAR-based forest biomass estimation[J]. J Nanjing For Univ (Nat Sci Ed), 2013, 37(3):163-169.DOI: 10.3969/j.issn.1000-2006.2013.03.029. | |
[6] | WANG K P, WANG T J, LIU X H. A review:individual tree species classification using integrated airborne LiDAR and optical imagery with a focus on the urban environment[J]. Forests, 2018, 10(1):1.DOI: 10.3390/f10010001. |
[7] | 黄华国. 林业定量遥感研究进展和展望[J]. 北京林业大学学报, 2019, 41(12):1-14. |
HUANG H G. Progress and perspective of quantitative remote sensing of forestry[J]. J Beijing For Univ, 2019, 41(12):1-14.DOI: 10.12171/j.1000-1522.20190326. | |
[8] | ASNER G P, MASCARO J, MULLER-LANDAU H C, et al. A universal airborne LiDAR approach for tropical forest carbon mapping[J]. Oecologia, 2012, 168(4):1147-1160.DOI: 10.1007/s00442-011-2165-z. |
[9] | DIAN Y Y, PANG Y, DONG Y F, et al. Urban tree species mapping using airborne LiDAR and hyperspectral data[J]. J Indian Soc Remote Sens, 2016, 44(4):595-603.DOI: 10.1007/s12524-015-0543-4. |
[10] | MAN Q X, DONG P L, YANG X M, et al. Automatic extraction of grasses and individual trees in urban areas based on airborne hyperspectral and LiDAR data[J]. Remote Sens, 2020, 12(17):2725.DOI: 10.3390/rs12172725. |
[11] | WALLACE L, LUCIEER A, WATSON C, et al. Development of a UAV-LiDAR system with application to forest inventory[J]. Remote Sens, 2012, 4(6):1519-1543.DOI: 10.3390/rs4061519. |
[12] | WALLACE L, MUSK R, LUCIEER A. An assessment of the repeatability of automatic forest inventory metrics derived from UAV-borne laser scanning data[J]. IEEE Trans Geosci Remote Sens, 2014, 52(11):7160-7169.DOI: 10.1109/TGRS.2014.2308208. |
[13] | 刘清旺, 李世明, 李增元, 等. 无人机激光雷达与摄影测量林业应用研究进展[J]. 林业科学, 2017, 53(7):134-148. |
LIU Q W, LI S M, LI Z Y, et al. Review on the applications of UAV-based LiDAR and photogrammetry in forestry[J]. Sci Silvae Sin, 2017, 53(7):134-148.DOI: 10.11707/j.1001-7488.20170714. | |
[14] | MICHEZ A, PIÉGAY H, JONATHAN L, et al. Mapping of riparian invasive species with supervised classification of Unmanned Aerial System (UAS) imagery[J]. Int J Appl Earth Obs Geoinf, 2016, 44:88-94.DOI: 10.1016/j.jag.2015.06.014. |
[15] | 陈向宇, 云挺, 薛联凤, 等. 基于激光雷达点云数据的树种分类[J]. 激光与光电子学进展, 2019, 56(12):203-214. |
CHEN X Y, YUN T, XUE L F, et al. Classification of tree species based on LiDAR point cloud data[J]. Laser Optoelectron Prog, 2019, 56(12):203-214.DOI: 10.3788/LOP56.122801. | |
[16] | GUAN H Y, YU Y T, JI Z, et al. Deep learning-based tree classification using mobile LiDAR data[J]. Remote Sens Lett, 2015, 6(11):864-873.DOI: 10.1080/2150704x.2015.1088668. |
[17] | 廖小罕, 肖青, 张颢. 无人机遥感:大众化与拓展应用发展趋势[J]. 遥感学报, 2019, 23(6):1046-1052. |
LIAO X H, XIAO Q, ZHANG H. UAV remote sensing:popularization and expand application development trend[J]. J Remote Sens, 2019, 23(6):1046-1052. | |
[18] | FRANKLIN S E, AHMED O S. Deciduous tree species classification using object-based analysis and machine learning with unmanned aerial vehicle multispectral data[J]. Int J Remote Sens, 2018, 39(15/16):5236-5245.DOI: 10.1080/01431161.2017.1363442. |
[19] | FENG X X, LI P J. A tree species mapping method from UAV images over urban area using similarity in tree-crown object histograms[J]. Remote Sens, 2019, 11(17):1982.DOI: 10.3390/rs11171982. |
[20] | SAARINEN N, VASTARANTA M, NÄSI R, et al. Assessing biodiversity in boreal forests with UAV-based photogrammetric point clouds and hyperspectral imaging[J]. Remote Sens, 2018, 10(2):338.DOI: 10.3390/rs10020338. |
[21] | ŠAŠAK J, GALLAY M, KANUK J, et al. Combined use of terrestrial laser scanning and UAV photogrammetry in mapping alpine terrain[J]. Remote Sens, 2019, 11(18):2154.DOI: 10.3390/rs11182154. |
[22] | TONG X H, LIU X F, CHEN P, et al. Integration of UAV-based photogrammetry and terrestrial laser scanning for the three-dimensional mapping and monitoring of open-pit mine areas[J]. Remote Sens, 2015, 7(6):6635-6662.DOI: 10.3390/rs70606635. |
[23] | ROŞCA S, SUOMALAINEN J, BARTHOLOMEUS H, et al. Comparing terrestrial laser scanning and unmanned aerial vehicle structure from motion to assess top of canopy structure in tropical forests[J]. Interface Focus, 2018, 8(2):20170038.DOI: 10.1098/rsfs.2017.0038. |
[24] | 张吴明, 李丹, 陈一铭, 等. 联合地基激光雷达与无人机摄影测量技术提取树高研究[J]. 北京师范大学学报(自然科学版), 2018, 54(6):764-771. |
ZHANG W M, LI D, CHEN Y M, et al. Integration between TLS and UAV photogrammetry techniques for retrieving tree height[J]. J Beijing Norm Univ (Nat Sci), 2018, 54(6):764-771.DOI: 10.16360/j.cnki.jbnuns.2018.06.011. | |
[25] | 曹明兰, 李亚东, 冯海英, 等. 倾斜摄影与激光扫描技术结合的3D森林景观建模[J]. 中南林业科技大学学报, 2019, 39(12):10-15,33. |
CAO M L, LI Y D, FENG H Y, et al. 3D forest landscape modeling with the combination of oblique photography and laser scanner technique[J]. J Cent South Univ For Technol, 2019, 39(12):10-15,33.DOI: 10.14067/j.cnki.1673-923x.2019.12.002. | |
[26] | 王楚虹, 刘浩然, 钟浩, 等. 联合UAV-LiDAR和HMLS技术的森林样地点云数据融合[J]. 中南林业科技大学学报, 2022, 42(3):26-38. |
WANG C H, LIU H R, ZHONG H, et al. Point cloud data fusion of forest plots based on UAV-LiDAR and HMLS technologies[J]. J Cent South Univ For Technol, 2022, 42(3):26-38.DOI: 10.14067/j.cnki.1673-923x.2022.03.004. | |
[27] | 黎华, 凯吾沙·塔依尔, 林木森, 等. 改进拟牛顿算法的点云稠密化应用研究[J]. 测绘科学, 2021, 46(12):83-90. |
LI H, Kai Wusha·Tayier, LIN M S, et al. Research on application of improved Quasi-Newton algorithm in point cloud densification[J]. Sci Surv Mapp, 2021, 46(12):83-90.DOI: 10.16251/j.cnki.1009-2307.2021.12.012. | |
[28] | TAO S L, WU F F, GUO Q H, et al. Segmenting tree crowns from terrestrial and mobile LiDAR data by exploring ecological theories[J]. ISPRS J Photogramm Remote Sens, 2015, 110:66-76.DOI: 10.1016/j.isprsjprs.2015.10.007. |
[29] | ZHANG Z Y, LIU X Y. Support vector machines for tree species identification using LiDAR-derived structure and intensity variables[J]. Geocarto Int, 2013, 28(4):364-378.DOI: 10.1080/10106049.2012.710653. |
[30] | ZIEGLER A, KÖNIG I R. Mining data with random forests:current options for real-world applications[J]. Wiley Interdiscip Rev Data Min Knowl Discov, 2014, 4(1):55-63.DOI: 10.1002/widm.1114. |
[31] | 杨龙, 孙中宇, 唐光良, 等. 基于微型无人机遥感的亚热带林冠物种识别[J]. 热带地理, 2016, 36(5):833-839. |
YANG L, SUN Z Y, TANG G L, et al. Identifying canopy species of subtropical forest by lightweight unmanned aerial vehicle remote sensing[J]. Trop Geogr, 2016, 36(5):833-839.DOI: 10.13284/j.cnki.rddl.002857. | |
[32] | 杜培军, 夏俊士, 薛朝辉, 等. 高光谱遥感影像分类研究进展[J]. 遥感学报, 2016, 20(2):236-256. |
DU P J, XIA J S, XUE Z H, et al. Review of hyperspectral remote sensing image classification[J]. J Remote Sens, 2016, 20(2):236-256.DOI: 10.11834/jrs.20165022. | |
[33] | 童庆禧, 张兵, 张立福. 中国高光谱遥感的前沿进展[J]. 遥感学报, 2016, 20(5):689-707. |
TONG Q X, ZHANG B, ZHANG L F. Current progress of hyperspectral remote sensing in China[J]. J Remote Sens, 2016, 20(5):689-707.DOI: 10.11834/jrs.20166264. | |
[34] | ZHONG H, LIN W S, LIU H R, et al. Identification of tree species based on the fusion of UAV hyperspectral image and LiDAR data in a coniferous and broad-leaved mixed forest in northeast China[J]. Front Plant Sci, 2022, 13:964769.DOI: 10.3389/fpls.2022.964769. |
[35] | 刘嘉政, 王雪峰, 王甜. 基于深度学习的树种图像自动识别[J]. 南京林业大学学报(自然科学版), 2020, 44(1):138-144. |
LIU J Z, WANG X F, WANG T. Automatic identification of tree species based on deep learning[J]. J Nanjing For Univ (Nat Sci Ed), 2020, 44(1):138-144.DOI: 10.3969/j.issn.1000-2006.201809004. |
[1] | 潘政尚, 马开森, 龙依, 赖珍贵, 孙华. 改进分类回归树模型的青冈枝叶点云分类研究[J]. 南京林业大学学报(自然科学版), 2024, 48(4): 123-131. |
[2] | 严宇, 李秋洁. 行道树靶标点云在线分割方法[J]. 南京林业大学学报(自然科学版), 2024, 48(4): 141-149. |
[3] | 高谢雨, 董利虎, 郝元朔. 基于TLS的抚育间伐对长白落叶松干形的影响[J]. 南京林业大学学报(自然科学版), 2023, 47(6): 85-94. |
[4] | 董瀚元, 于颖, 范文义. 星载激光雷达GEDI数据林下地形反演性能验证[J]. 南京林业大学学报(自然科学版), 2023, 47(2): 141-149. |
[5] | 王帆, 贾炜玮, 唐依人, 李丹丹. 基于TLS的红松树冠半径提取及其外轮廓模型构建[J]. 南京林业大学学报(自然科学版), 2023, 47(1): 13-22. |
[6] | 沈文娟, 纪梅, 李明诗. 基于多源遥感的森林变化对区域温度影响的监测方法研究进展[J]. 南京林业大学学报(自然科学版), 2022, 46(3): 1-11. |
[7] | 巨一琳, 姬永杰, 黄继茂, 张王菲. 联合LiDAR和多光谱数据森林地上生物量反演研究[J]. 南京林业大学学报(自然科学版), 2022, 46(1): 58-68. |
[8] | 赵颖慧, 郭新龙, 甄贞. 基于光学-ALS变量组合和非参数模型的天然次生林地上生物量估算[J]. 南京林业大学学报(自然科学版), 2021, 45(4): 49-57. |
[9] | 花伟成, 田佳榕, 孙心雨, 徐雁南. 基于TLS数据的杨树削度方程建立及材积估算[J]. 南京林业大学学报(自然科学版), 2021, 45(4): 41-48. |
[10] | 赵颖慧, 杨海城, 甄贞. 基于ULS、TLS和超声测高仪的天然次生林中不同林冠层树高估测[J]. 南京林业大学学报(自然科学版), 2021, 45(4): 23-32. |
[11] | 蒋佳文, 温小荣, 顾海波, 张峥男, 刘方舟, 张严利, 孙圆. 基于多站扫描的点云特征参数与材积结构动态分析[J]. 南京林业大学学报(自然科学版), 2019, 43(6): 83-90. |
[12] | 赵颖慧, 张大力, 甄贞. 基于非参数分类算法和多源遥感数据的单木树种分类[J]. 南京林业大学学报(自然科学版), 2019, 43(5): 103-112. |
[13] | 陶欢,李存军,谢春春,周静平,淮贺举,蒋丽雅,李凤涛. 基于HSV阈值法的无人机影像变色松树识别[J]. 南京林业大学学报(自然科学版), 2019, 43(03): 99-106. |
[14] | 邱赛,邢艳秋,田静,丁建华. 基于ICESat-GLAS波形数据估测森林郁闭度[J]. 南京林业大学学报(自然科学版), 2016, 40(05): 99-106. |
[15] | 张天安,云挺,薛联凤,安锋. 基于地面激光雷达的活立木枝干三维建模[J]. 南京林业大学学报(自然科学版), 2015, 39(04): 163-167. |
阅读次数 | ||||||
全文 |
|
|||||
摘要 |
|
|||||