基于ICESat-GLAS数据和模糊模式识别算法识别森林类型

doi:10.12302/j.issn.1000-2006.202003008

国家林草科技领军期刊
中国精品科技期刊
中国高校百佳科技期刊
江苏省新闻出版政府奖期刊奖
RCCSE林学权威期刊（A+）
CSCD核心期刊
Scopus数据库收录期刊
中文核心期刊
SCD核心期刊

作者加群：102861116

微信公众号：南京林业大学学报

高级检索

南京林业大学学报（自然科学版） ›› 2021, Vol. 45 ›› Issue (4): 33-40.doi: 10.12302/j.issn.1000-2006.202003008

• 专题报道I (执行主编李凤日) • 上一篇下一篇

基于ICESat-GLAS数据和模糊模式识别算法识别森林类型

蔡龙涛(), 邢涛^*(), 邢艳秋, 丁建华, 黄佳鹏, 崔阳, 秦磊

东北林业大学工程技术学院, 黑龙江哈尔滨 150040

收稿日期:2020-03-03 接受日期:2021-03-22 出版日期:2021-07-30 发布日期:2021-07-30
通讯作者: 邢涛
基金资助:
国家重点研发计划(2017YFD060090402);中央高校基本科研业务费专项资金项目(2572019AB18);卫星测绘技术与应用国家测绘地理信息局重点实验室项目(KLSMTA-201706)

Identification of forest types based on ICESat-GLAS data and fuzzy pattern recognition algorithm

CAI Longtao(), XING Tao^*(), XING Yanqiu, DING Jianhua, HUANG Jiapeng, CUI Yang, QIN Lei

College of Engineering and Technology, Northeast Forestry University, Harbin 150040, China

Received:2020-03-03 Accepted:2021-03-22 Online:2021-07-30 Published:2021-07-30
Contact: XING Tao

摘要/Abstract

摘要：

【目的】利用冰、云和陆地高程卫星-地球科学激光测高系统(ICESat-GLAS)回波波形数据,通过模糊模式识别算法,提出波形特征参数组合,对森林类型进行识别研究,以期提高森林类型分类精度。【方法】利用不同森林类型冠层在GLAS回波波形上表现出的差异性,提取波形特征参数 $R fi t 1$ 、K₁'和 $K 1 ¯$ ;将本研究提取的波形特征参数与其他波形特征参数相结合,建立波形特征参数组合;对样本数据波形特征参数进行指标归一化和奇异点检测处理,剔除样本数据中的奇异点样本;结合模糊模式识别算法,计算不同森林类型分类精度。【结果】针叶林和阔叶林森林类型分类总精度为96.30%,其中,针叶林和阔叶林森林类型分类精度分别为92.86%和97.50%;针叶林、阔叶林和混交林森林类型分类总精度为84.51%,其中,针叶林、阔叶林和混交林森林类型分类精度分别为85.71%、97.50%和52.94%。【结论】模糊模式识别算法在森林类型分类方面具有一定优势,尤其在针叶林和阔叶林森林类型识别方面,识别精度较高。

关键词: 地球科学激光测高系统, 森林类型, 模糊模式识别, 波形特征参数, 奇异点, 隶属度

Abstract:

【Objective】 Using space-borne LiDAR waveform data and the fuzzy pattern recognition algorithm, combined with the waveform characteristic parameters proposed in this paper, the forest type was identified and studied to improve the accuracy of forest type classification. 【Method】 First, using the differences in GLAS-received waveforms of the canopies of different forest types, the waveform characteristic parameters $R fi t 1$ , $K 1'$ and $K 1 ¯$ were acquired. Second, the extracted waveform feature parameters were combined with the other waveform feature parameters to establish a combination of waveform feature parameters. Thereafter, the waveform feature parameters of the sample data were subjected to index normalization and singularity detection processing to unify the dimensions of different waveform feature parameters and remove the sample data. Finally, combined with the fuzzy pattern recognition algorithm, the classification accuracy of different forest types was calculated. 【Result】 The total classification accuracies of coniferous and broadleaved forest types were 96.30%, of which the classification accuracies of coniferous and broadleaved forest types were 92.86% and 97.50%, respectively. Furthermore, the total accuracy was 84.51%, of which the classification accuracies of coniferous forest, broadleaved forest, and mixed forest types were 85.71%, 97.50% and 52.94%, respectively. 【Conclusion】 The fuzzy pattern recognition algorithm has certain advantages for the forest-type classification. This is especially so in the classification of coniferous and broadleaved forests, where the classification accuracy was higher.

Key words: geoscience laser altimeter system (GLAS), forest type, fuzzy pattern recognition, waveform characteristic parameters, singular point, degree of membership

中图分类号:

S771:S758

蔡龙涛,邢涛,邢艳秋,等. 基于ICESat-GLAS数据和模糊模式识别算法识别森林类型[J]. 南京林业大学学报（自然科学版）, 2021, 45(4): 33-40.

CAI Longtao, XING Tao, XING Yanqiu, DING Jianhua, HUANG Jiapeng, CUI Yang, QIN Lei. Identification of forest types based on ICESat-GLAS data and fuzzy pattern recognition algorithm[J].Journal of Nanjing Forestry University (Natural Science Edition), 2021, 45(4): 33-40.DOI: 10.12302/j.issn.1000-2006.202003008.

图/表 5

图1

图2

表1

不同参数组合针叶林和阔叶林森林类型分类结果"

特征参数组合 character metrics combination	分类总数 total number of types	正确分类数 correct classification number	分类总精度/% accuracy of classification	Kappa系数 Kappa coefficient
$R fi t 1$ 、K₁'、 $K 1 ¯$	54	50	92.59	0.820 3
AGS、SGS、MSGS	54	49	90.74	0.778 0
$R fi t 1$ 、K₁'、 $K 1 ¯$ 、AGS、SGS、MSGS	54	52	96.30	0.903 6

表1

表2

表3

参考文献 28

[1]	FIELD C B, RAUPACH M R, EBRARY I, et al. The global carbon cycle: integrating humans, climate, and the natural world[M]. Chicago: Island Press, 2004.
[2]	汤旭光. 基于激光雷达与多光谱遥感数据的森林地上生物量反演研究[D]. 北京:中国科学院研究生院(东北地理与农业生态研究所), 2013.
	TANG X G. Estimation of forest aboveground biomass by integrating ICESat/GLAS waveform and TM data[D]. Beijing: Graduate School of Chinese Academy of Sciences (Northeast Institute of Geography and Agricultural Ecology), 2013.
[3]	邱赛. ICESat-GLAS波形与HJ-1A高光谱影像联合反演森林地上生物量的研究[D]. 哈尔滨:东北林业大学, 2016.
	QIU S. The research of regional forest above ground biomass inversion combining ICESat-GLAS waveform and HJ-1A hyperspectral imageries[D]. Harbin:Northeast Forestry University, 2016.
[4]	吴文跃, 姚顺彬, 徐志扬. 基于森林资源清查数据的江西省主要森林类型净生产力研究[J]. 南京林业大学学报(自然科学版), 2019, 43(5):193-198.
	WU W Y, YAO S B, XU Z Y, et al. A study on the net productivity of main forest types in Jiangxi Province based on forest resources inventory data[J]. J Nanjing For Univ (Nat Sci Ed), 2019, 43(5):193-198. DOI: 10.3969/j.issn.1000-2006.201903045. doi: 10.3969/j.issn.1000-2006.201903045
[5]	MOHOLDT G, NUTH C, HAGEN J O, et al. Recent elevation changes of Svalbard glaciers derived from ICESat laser altimetry[J]. Remote Sens Environ, 2010, 114(11):2756-2767.DOI: 10.1016/j.rse.2010.06.008. doi: 10.1016/j.rse.2010.06.008
[6]	CARABAJAL C C, HARDING D J, SUCHDEO V P, et al. Development of AN ICESat geodetic control database and evaluation of global topographic assets[J]. American Geophysical Union, Fall Meeting, 2010(6):1907-1910. DOI: 10.1038/nmeth.2835. doi: 10.1038/nmeth.2835
[7]	LEFSKY M A. A global forest canopy height map from the moderate resolution imaging spectroradiometer and the geoscience laser altimeter system[J]. Geophys Res Lett, 2010, 37(15):L15401.DOI: 10.1029/2010GL043622. doi: 10.1029/2010GL043622
[8]	KHALEFA E, SMIT I P J, NICKLESS A, et al. Retrieval of savanna vegetation canopy height from ICESat-GLAS spaceborne LiDAR with terrain correction[J]. IEEE Geosci Remote Sens Lett, 2013, 10(6):1439-1443.DOI: 10.1109/LGRS.2013.2259793. doi: 10.1109/LGRS.2013.2259793
[9]	XING Y Q, DE GIER A, ZHANG J J, et al. An improved method for estimating forest canopy height using ICESat-GLAS full waveform data over sloping terrain:a case study in Changbai Mountains,China[J]. Int J Appl Earth Obs Geoinformation, 2010, 12(5):385-392.DOI: 10.1016/j.jag.2010.04.010. doi: 10.1016/j.jag.2010.04.010
[10]	WANG C, TANG F X, LI L W, et al. Wavelet analysis for ICESat/GLAS waveform decomposition and its application in average tree height estimation[J]. IEEE Geosci Remote Sens Lett, 2013, 10(1):115-119.DOI: 10.1109/LGRS.2012.2194692. doi: 10.1109/LGRS.2012.2194692
[11]	邱赛, 邢艳秋, 田静, 等. 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(5):99-106.
	QIU S, XING Y Q, TIAN J, et al. Estimation of forest canopy density based on ICESat-GLAS waveform data[J]. J Nanjing For Univ (Nat Sci Ed), 2016, 40(5):99-106.DOI: 10.3969/j.issn.1000-2006.2016.05.016. doi: 10.3969/j.issn.1000-2006.2016.05.016
[12]	XI X H, HAN T T, WANG C, et al. Forest above ground biomass inversion by fusing GLAS with optical remote sensing data[J]. ISPRS Int J Geo-Inf, 2016, 5(4):45.DOI: 10.3390/ijgi5040045. doi: 10.3390/ijgi5040045
[13]	WANG Y, NI W J, SUN G Q, et al. Slope-adaptive waveform metrics of large footprint lidar for estimation of forest aboveground biomass[J]. Remote Sens Environ, 2019, 224:386-400.DOI: 10.1016/j.rse.2019.02.017. doi: 10.1016/j.rse.2019.02.017
[14]	FATOYINBO T E, SIMARD M. Height and biomass of mangroves in Africa from ICESat/GLAS and SRTM[J]. Int J Remote Sens, 2013, 34(2):668-681.DOI: 10.1080/01431161.2012.712224. doi: 10.1080/01431161.2012.712224
[15]	RANSON K J, SUN G, KOVACS K, et al. Landcover attributes from ICESat GLAS data in Central Siberia[C]// IGARSS 2004.2004 IEEE International Geoscience and Remote Sensing Symposium.September 20-24,2004,Anchorage,AK,USA.IEEE, 2004:753-756.DOI: 10.1109/IGARSS.2004.1368511. doi: 10.1109/IGARSS.2004.1368511
[16]	ZHANG J J, DE GIER A, XING Y Q, et al. Full waveform-based analysis for forest type information derivation from large footprint spaceborne lidar data[J]. Photogramm Eng Remote Sensing, 2011, 77(3):281-290.DOI: 10.14358/pers.77.3.281. doi: 10.14358/pers.77.3.281
[17]	LI L C, XING Y Q. Study on forest type classification based on ICESat-GLAS lidar data[C]// 2011 2nd International Conference on Artificial Intelligence,Management Science and Electronic Commerce (AIMSEC).August 8-10,2011,Deng Feng,China.IEEE, 2011:3379-3382.DOI: 10.1109/AIMSEC.2011.6011388. doi: 10.1109/AIMSEC.2011.6011388
[18]	刘美爽, 邢艳秋, 李立存, 等. 基于星载激光雷达数据和支持向量分类机方法的森林类型识别[J]. 东北林业大学学报, 2014, 42(2):124-128.
	LIU M S, XING Y Q, LI L C, et al. Forest type identification with spaceborne LIDAR data and C-support vector classification[J]. J Northeast For Univ, 2014, 42(2):124-128.DOI: 10.13759/j.cnki.dlxb.2014.02.030. doi: 10.13759/j.cnki.dlxb.2014.02.030
[19]	蔡龙涛, 邢艳秋, 黄佳鹏, 等 . 基于 ICESat-GLAS 数据和回波仿真原理识别森林类型[J]. 中南林业科技大学学报, 2021, 41(1):60-68.
	CAI L T, XING Y Q, HUANG J P, et al. Forest types identification based on ICESat-GLAS data and echo simulation principles[J]. J Central South Univ For Technol, 2021, 41(1):60-68.
[20]	WU H B, XING Y Q. Wavelet transform and its application to ICESat-GLAS full waveform data[C]// 2010 International Symposium on Intelligence Information Processing and Trusted Computing.October 28-29,2010,Huanggang,China.IEEE, 2010:373-377.DOI: 10.1109/IPTC.2010.38. doi: 10.1109/IPTC.2010.38
[21]	邱赛, 邢艳秋, 李立存, 等. 基于小波变换的ICESAT-GlAS波形处理[J]. 森林工程, 2012, 28(5):33-35,59.
	QIU S, XING Y Q, LI L C, et al. ICESAT-GLAS data processing based on wavelet transform[J]. For Eng, 2012, 28(5):33-35,59.DOI: 10.16270/j.cnki.slgc.2012.05.022. doi: 10.16270/j.cnki.slgc.2012.05.022
[22]	LI D, XU L J, LI X L, et al. Full-waveform LiDAR signal filtering based on Empirical Mode Decomposition method[C]// 2013 IEEE International Geoscience and Remote Sensing Symposium-IGARSS.July 21-26,2013,Melbourne,VIC,Australia.IEEE, 2013:3399-3402.DOI: 10.1109/IGARSS.2013.6723558. doi: 10.1109/IGARSS.2013.6723558
[23]	YANG T, WANG C, LI G C, et al. Forest canopy height mapping over China using GLAS and MODIS data[J]. Sci China Earth Sci, 2015, 58(1):96-105.DOI: 10.1007/s11430-014-4905-5. doi: 10.1007/s11430-014-4905-5
[24]	MAHONEY C, KLJUN N, LOS S, et al. Slope estimation from ICESat/GLAS[J]. Remote Sens, 2014, 6(10):10051-10069.DOI: 10.3390/rs61010051. doi: 10.3390/rs61010051
[25]	PANG Y, LEFSKY M, SUN G Q, et al. Impact of footprint diameter and off-nadir pointing on the precision of canopy height estimates from spaceborne lidar[J]. Remote Sens Environ, 2011, 115(11):2798-2809.DOI: 10.1016/j.rse.2010.08.025. doi: 10.1016/j.rse.2010.08.025
[26]	汪泉, 宋文龙, 张怡卓, 等. 基于改进VGG16 网络的机载高光谱针叶树种分类研究[J]. 森林工程, 2021, 37(3):79-87.
	WANG Q, SONG W L, ZHANG Y Z, et al. Study on hyperspectral conifer species classification based on improved VGG16 network[J]. For Eng, 2021, 37(3):79-87.
[27]	DUNCANSON L I, NIEMANN K O, WULDER M A. Estimating forest canopy height and terrain relief from GLAS waveform metrics[J]. Remote Sens Environ, 2010, 114(1):138-154.DOI: 10.1016/j.rse.2009.08.018. doi: 10.1016/j.rse.2009.08.018
[28]	JIHYUN H A, SEOK S, LEE J S. Robust outlier detection using the instability factor[J]. Knowl-Based Syst, 2014, 63:15-23.DOI: 10.1016/j.knosys.2014.03.001. doi: 10.1016/j.knosys.2014.03.001

森林类型 forest type	分类结果 result of classification		类型总数 total number of types	分类精度/% accuracy of classification
森林类型 forest type	针叶林 coniferous forest	阔叶林 broadleaved forest	类型总数 total number of types	分类精度/% accuracy of classification
针叶林 coniferous forest	13	1	14	92.86
阔叶林 broadleaved forest	1	39	40	97.50

森林类型 forest type	分类结果 result of classification			类型总数 total number of types	分类精度/% accuracy of classification	分类总精度/% accuracy of total classification	Kappa系数 Kappa coefficient
森林类型 forest type	针叶林 coniferous forest	阔叶林 broadleaved forest	混交林 mixed forest	类型总数 total number of types	分类精度/% accuracy of classification	分类总精度/% accuracy of total classification	Kappa系数 Kappa coefficient
针叶林 coniferous forest	12	1	1	14	85.71
阔叶林 broadleaved forest	1	39	0	40	97.50	84.51	0.725 7
混交林 mixed forest	3	5	9	17	52.94

基于ICESat-GLAS数据和模糊模式识别算法识别森林类型

Identification of forest types based on ICESat-GLAS data and fuzzy pattern recognition algorithm

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 5

参考文献 28

相关文章 9

编辑推荐

Metrics

本文评价

作者

读者

审稿

[1]	朱家琪, 满秀玲, 王飞. 寒温带4种森林类型土壤团聚体有机碳氮特征[J]. 南京林业大学学报（自然科学版）, 2021, 45(5): 71-83.
[2]	吴国训,唐学君,阮宏华,罗细芳. 基于森林资源清查的江西省森林碳储量及固碳潜力研究[J]. 南京林业大学学报（自然科学版）, 2019, 43(01): 105-110.
[3]	魏强,张广忠,凌雷,柴春山. 甘肃兴隆山主要森林类型凋落物及土壤层的蓄水功能[J]. 南京林业大学学报（自然科学版）, 2013, 37(02): 78-84.
[4]	陶宝先,张金池. 南京地区主要森林类型空气负离子变化特征[J]. 南京林业大学学报（自然科学版）, 2012, 36(02): 147-150.
[5]	伊力塔，严晓素，余树全，包春泉，屠永海，潘昌尧. 浙江省不同森林类型林分健康指标体系[J]. 南京林业大学学报（自然科学版）, 2012, 36(01): 145-148.
[6]	王登峰，杨志刚，魏安世. 纹理信息在遥感影像分类中的应用[J]. 南京林业大学学报（自然科学版）, 2010, 34(03): 97-100.
[7]	蒋文伟;姜志林;余树全;刘安兴;钱新标. 安吉主要森林类型水源涵养功能的分析与评价[J]. 南京林业大学学报（自然科学版）, 2002, 26(04): 71-74.
[8]	徐世永;林杰;陈涛. 数字化CIR图像专题特征的提取及成图[J]. 南京林业大学学报（自然科学版）, 2000, 24(03): 49-52.
[9]	刘昉勋;黄致远. 关于森林分类问题的探讨[J]. 南京林业大学学报（自然科学版）, 1985, 9(01): 89-96.