基于改进的局部最大值法提取杉木单木位置

doi:10.13466/j.cnki.lyzygl.2022.05.008

摘要/Abstract

摘要：

杉木是我国重要的用材树种之一,开展杉木单木位置和株数密度提取研究对调控其林分空间结构和功能、提高林分质量具有重要作用。基于无人机遥感影像,以浙江龙泉市杉木纯林为研究对象,采用改进的局部最大值法提取杉木单木位置和株数,并与参考株数进行对比分析。改进的局部最大值方法的采样间隔参数对单木位置和株数提取精度起到重要影响。在合适的采样间隔参数下:密和疏两种郁闭度样地单木位置提取总体精度分别为82.10%和80.17%、错分误差分别为24.12%和18.18%、漏分误差分别为17.90%和19.83%;密和疏两种郁闭度样地的监测株数和参考株数都十分相近,相对精度分别为93.77%和98.35%;林分株数密度与总体精度和错分误差呈负相关,与漏分误差呈正相关。改进的局部最大值方法能够较准确地提取不同郁闭度的杉木单木位置和株数,为智能、快速、准确地提取杉木单木位置和株数提供了一种可行的方法。

关键词: 局部最大值法, 无人机, 单木, 株数

Abstract:

Chinese fir is one of the important timber species in China.The extraction of individual tree position and density plays an important role in regulating its stand spatial structure and function,and improving stand quality.Based on the unmanned aerial vehicle(UAV)remote sensing image,taking the Chinese fir pure forest in Longquan City of Lishui as the research object,the improved local maximum method was used to extract the crown position and number of Chinese fir trees,which was compared and analyzed with the measured number of trees.The results show that the sampling interval parameter of the improved local maximum method had an important influence on the extraction accuracy of individual tree number.Under the appropriate sampling interval parameters,the overall accuracy of individual tree position detection for sample plots with dense and sparse tree density were 82.10% and 80.17%,commission errors were 24.12% and 18.18%,and omission errors were 17.90% and 19.83%,respectively.The detected tree density and measured tree density for sample plots with dense and sparse tree density were very similar,and the accuracy was 93.77% and 98.35%,respectively.The tree density had a negative correlation with the overall accuracy and commission error,and a positive correlation with the omission error.The improved local maximum method can accurately extract the number of individual trees of Chinese fir forest with different tree density,which provides a feasible method for intelligent,fast and accurate extraction of individual crown position and tree density of Chinese fir.

Key words: local maximum method, UAV, individual tree, tree density

中图分类号:

P237
S791.27

李苏春, 林露花, 夏磊, 胡璐璐, 徐小军. 基于改进的局部最大值法提取杉木单木位置[J]. 林业资源管理, 2022, 0(5): 60-68.

LI Suchun, LIN Luhua, XIA Lei, HU Lulu, XU Xiaojun. Extracting Individual Tree Position of Chinese Fir Based on an Improved Local Maximum Method[J]. FOREST RESOURCES WANAGEMENT, 2022, 0(5): 60-68.

图/表 10

图1

表1

图2

图3

图4

图5

表2

表3

图6

图7

参考文献 31

[1]	国家林业局. 中国森林资源报告[M]. 北京: 中国林业出版社, 2014.
[2]	张英凯, 刘鹏举, 刘长春, 等. 基于空间聚类的杉木生长预测方法[J]. 林业科学, 2019, 55(11):137-144.
[3]	Zhang Yanhua, Tian Ye, Ding Sihui, et al. Growthcarbon storage, and optimal rotation in poplar plantations:a case study on clone and planting spacing effects[J]. Forests, 2020, 11(8):842. doi: 10.3390/f11080842
[4]	李晨, 李佳, 王明果, 等. 无人机RGB影像中人工林单木位置的提取[J]. 林业工程学报, 2021, 6(2):141-147.
[5]	庞丽峰, 陆元昌, 洪玲霞, 等. 森林经营单位的单木定位技术研究及其应用[J]. 林业资源管理, 2014(2):46-51.
[6]	宋亚斌, 江腾宇, 彭检贵, 等. 基于单木位置的杉木林分空间竞争关系研究[J]. 湖北林业科技, 2019, 48(6):16-19.
[7]	Komura R, Muramoto K I. Classification of forest stand considering shapes and sizes of tree crown calculated from high spatial resolution satellite image[C]// IEEE International Geoscience and Remote Sensing Symposium.IEEE, 2007:4356-4359.
[8]	Ferreira M P, Almeida D R A D, Papa D D A, et al. Individual tree detection and species classification of Amazonian palms using UAV images and deep learning[J]. Forest Ecology and Management, 2020, 475, 118397. doi: 10.1016/j.foreco.2020.118397
[9]	Forzieri G, Guarnieri L, Vivoni E R, et al. Multiple attribute decision making for individual tree detection using high-resolution laser scanning[J]. Forest Ecology and Management, 2009, 258(11):2501-2510. doi: 10.1016/j.foreco.2009.09.006
[10]	Li Xiang, Zhao Yinghui, et al. Suitable model of detecting the position of individual treetop based on local maximum method[J]. Journal of Beijing Forestry University, 2015, 37(3):27-33.
[11]	孙钊, 潘磊, 孙玉军. 基于无人机影像的高郁闭度杉木纯林树冠参数提取[J]. 北京林业大学学报, 2020, 42(10):20-26.
[12]	Hu Baoxin, Li Jili, Jing Linhai, et al. Improving the efficiency and accuracy of individual tree crown delineation from high-density LiDAR data[J]. International Journal of Applied Earth Observation and Geoinformation, 2014, 26:145-155. doi: 10.1016/j.jag.2013.06.003
[13]	王娟, 张超, 陈巧, 等. 结合无人机可见光和激光雷达数据的杉木树冠信息提取[J]. 西南林业大学学报:自然科学, 2022, 42(1):133-141.
[14]	Lin Yi, Jiang Miao, Yao Yunjun, et al. Use of UAV oblique imaging for the detection of individual trees in residential environments[J]. Urban forestry & urban greening, 2015, 14(2):404-412.
[15]	Qin Haiming, Zhou Weiqi, Yao Yang, et al. Individual tree segmentation and tree species classification in subtropical broadleaf forests using UAV-based LiDAR,hyperspectral,and ultrahigh-resolution RGB data[J]. Remote Sensing of Environment, 2022, 280,113143. doi: 10.1016/j.rse.2022.113143
[16]	Gomes M F, Maillard P, Deng H. Individual tree crown detection in sub-meter satellite imagery using Marked Point Processes and a geometrical-optical model[J]. Remote Sensing of Environment, 2018, 211:184-195. doi: 10.1016/j.rse.2018.04.002
[17]	陈时跃. 基于无人机可见光影像的郁闭山地杉木人工林的单木与林分结构参数提取[D]. 杭州: 浙江农林大学, 2020.
[18]	Ke Y, Quackenbush L J. A review of methods for automatic individual tree-crown detection and delineation from passive remote sensing[J]. International Journal of Remote Sensing, 2011, 32(17):4725-4747. doi: 10.1080/01431161.2010.494184
[19]	Kukko A, Kaijaluoto R, Kaartinen H, et al. Graph SLAM correction for single scanner MLS forest data under boreal forest canopy[J]. ISPRS journal of photogrammetry and remote sensing, 2017, 132:199-209. doi: 10.1016/j.isprsjprs.2017.09.006
[20]	Stereńczak K, Kraszewski B, Mielcarek M, et al. Mapping individual trees with airborne laser scanning data in an European lowland forest using a self-calibration algorithm[J]. International Journal of Applied Earth Observation and Geoinformation, 2020, 93, 102191. doi: 10.1016/j.jag.2020.102191
[21]	Mohan M, Mendonça B A F D, Silva C A, et al. Optimizing individual tree detection accuracy and measuring forest uniformity in coconut(Cocos nucifera L.) plantations using airborne laser scanning[J]. Ecological Modelling, 2019, 409:108736. doi: 10.1016/j.ecolmodel.2019.108736
[22]	Pouliot D A, King D J, Bell F W, et al. Automated tree crown detection and delineation in high-resolution digital camera imagery of coniferous forest regeneration[J]. Remote sensing of environment, 2002, 82(2):322-334. doi: 10.1016/S0034-4257(02)00050-0
[23]	Lisiewicz M, Kamińska A, Stereńczak K. Recognition of specified errors of Individual Tree Detection methods based on Canopy Height Model[J]. Remote Sensing Applications:Society and Environment, 2022, 25, 100690. doi: 10.1016/j.rsase.2021.100690
[24]	彭后生, 李晨晨, 石峤瑀. 林木资源分类估价及其基准价研究——以浙江省龙泉市为例[J]. 林业资源管理, 2020,(6):61-65.
[25]	董新宇, 李家国, 陈瀚阅, 等. 无人机遥感影像林地单株立木信息提取[J]. 遥感学报, 2019, 23(6):1269-1280.
[26]	Dralle K, Rudemo M. Automatic estimation of individual tree positions from aerial photos[J]. Canadian Journal of Forest Research, 1997, 27(11):1728-1736. doi: 10.1139/x97-130
[27]	Xu Xiaojun, Zhou Zhongsheng, Tang Yan, et al. Individual tree crown detection from high spatial resolution imagery using a revised local maximum filtering[J]. Remote Sensing of Environment, 2021, 258(8), 112397. doi: 10.1016/j.rse.2021.112397
[28]	金忠明, 曹姗姗, 王蕾, 等. 基于U-Net和分水岭算法的无人机单木树冠提取方法[J]. 西北林学院学报, 2020, 35(6):194-204.
[29]	李文静, 王瑞瑞, 石伟, 等. 基于无人机多光谱影像的单木树冠提取方法[J]. 福建农林大学学报:自然科学版, 2020, 49(5):639-645.
[30]	袁梓健, 舒清态, 赵洪莹, 等. 基于无人机可见光影像的华山松人工林计测参数提取[J]. 林业资源管理, 2021(2):110-116.
[31]	胡馨月, 倪海明, 戚大伟. 基于无人机影像的树木株数提取[J]. 森林工程, 2021, 37(1):6-12.

样方号	样地1			样地2			样方号	样地1			样地2
样方号	面积/ m²	株数/ 株	密度/ (株/hm²)	面积/ m²	株数/ 株	密度/ (株/hm²)	样方号	面积/ m²	株数/ 株	密度/ (株/hm²)	面积/ m²	株数/ 株	密度/ (株/hm²)
1	121.2	37	3053	120.6	17	1410	6	70.2	14	1994	89.5	13	1453
2	99.9	27	2703	85.2	10	1174	7	86.9	32	3683	118.6	15	1265
3	94.0	32	3404	122.9	11	895	8	76.7	28	3651	102.9	15	1458
4	72.4	35	4834	105.4	10	949	9	88.9	28	3150	98.0	21	2143
5	64.0	24	3750	77.3	9	1164	整体	774.2	257	3320	920.4	121	1315

样地	参考株数	监测株数	正确株数	错分株数	漏分株数	OA/%	CE/%	OE/%
样方1	37	53	36	17	1	97.30	45.95	2.70
样方2	27	29	23	6	4	85.19	22.22	14.81
样方3	32	29	24	5	8	75.00	15.63	25.00
样方4	35	28	25	3	10	71.43	8.57	28.57
样方5	24	21	18	3	6	75.00	12.50	25.00
样方6	14	18	12	6	2	85.71	42.86	14.29
样方7	32	32	25	7	7	78.13	21.88	21.88
样方8	28	31	23	8	5	82.14	28.57	17.86
样方9	28	32	25	7	3	89.29	25.00	10.71
整个样地	257	273	211	62	46	82.10	24.12	17.90

样地	参考株数	监测株数	正确株数	错分株数	漏分株数	OA/%	CE/%	OE/%
样方1	17	18	13	5	4	76.47	29.41	23.53
样方2	10	12	8	4	2	80.00	40.00	20.00
样方3	11	11	8	3	3	72.73	27.27	27.27
样方4	10	10	9	1	1	90.00	10.00	10.00
样方5	9	7	7	0	2	77.78	0.00	22.22
样方6	13	13	11	2	2	84.62	15.38	15.38
样方7	15	14	12	2	3	80.00	13.33	20.00
样方8	15	17	14	3	1	93.33	20.00	6.67
样方9	21	17	15	2	6	71.43	9.52	28.57
整个样地	121	119	97	22	24	80.17	18.18	19.83