[1]王广科,吴达胜,方陆明.基于多源数据的毛竹林信息自动提取[J].浙江林业科技,2023,43(03):79-86.[doi:10.3969/j.issn.1001-3776.2023.03.011]
 WANG Guang-ke,WU Da-sheng,FANG Lu-ming.Information Extraction of Phyllostachys edulis Forest based on Multi Source Data[J].Journal of Zhejiang Forestry Science and Technology,2023,43(03):79-86.[doi:10.3969/j.issn.1001-3776.2023.03.011]
点击复制

基于多源数据的毛竹林信息自动提取()
分享到:

《浙江林业科技》[ISSN:1001-3776/CN:33-1112/S]

卷:
43
期数:
2023年03期
页码:
79-86
栏目:
出版日期:
2023-05-20

文章信息/Info

Title:
Information Extraction of Phyllostachys edulis Forest based on Multi Source Data
文章编号:
1001-3776(2023)03-0079-08
作者:
王广科1吴达胜123方陆明123
1.浙江农林大学 数学与计算机科学学院,浙江 杭州 311300;2.浙江省林业智能监测与信息技术研究重点实验室,浙江 杭州 311300;3.林业感知技术与智能装备国家林业和草原局重点实验室,浙江 杭州 311300
Author(s):
WANG Guang-ke1WU Da-sheng123FANG Lu-ming123
1.School of Mathematics and Computer Science, Zhejiang A & F University, Hangzhou 311300, China;2.Key Laboratory of Forestry Intelligent Monitoring and Information Technology of Zhejiang Province, Hangzhou 311300, China;3.Key Laboratory of State Forestry and Grassland Administration on Forestry Sensing Technology and Intelligent Equipment, Hangzhou 311300, China
关键词:
毛竹林遥感机器学习信息提取
Keywords:
Phyllostachys edulis forest machine learning remote sensing information extraction
分类号:
S711
DOI:
10.3969/j.issn.1001-3776.2023.03.011
文献标志码:
A
摘要:
毛竹Phyllostachysedulis是重要的经济林种,快速准确地获取毛竹林的面积及郁闭度等信息可对毛竹林的高效经营管理提供巨大帮助。基于人工样地的森林资源调查耗时费力且效率低下,故利用遥感图像等较低成本的数据源估测大范围毛竹林的面积及其郁闭度等信息具有重要意义。本文以浙江省安吉县为研究区域,基于Sentinel-2中等分辨率遥感图像、数字高程模型(DigitalElevationModel,DEM)数据及森林资源二类调查数据,利用CatBoost、随机森林(RandomForest,RF)、支持向量机(SupportVectorMachine,SVM)三种算法,通过二分类方法估算毛竹林面积和多分类方法估算毛竹林郁闭度。结果表明:在估算毛竹林面积时,CatBoost、RF、SVM的总体分类精度分别为97.75%、96.54%、95.05%,Kappa系数分别为0.96、0.93、0.90;在估算毛竹林郁闭度时,以上三种算法的总体分类精度分别为73.38%、73.49%、69.9%,Kappa系数分别为0.52、0.52、0.45。在本文研究数据中,0.7、0.8郁闭度样本(大类别样本)占样本总数的89.42%,剩余类别样本(小类别样本)只占样本总数的10.58%。样本失衡是导致毛竹林郁闭度估算精度严重偏低的重要原因。将样本分成两种类别后(占比较大的样本和占比较小的样本)分别进行建模,模型估测效果有了较大改善,各模型郁闭度估算精度均在78%以上,其中以CatBoost最优,总体精度达到83.45%。基于Sentinel-2遥感图像、DEM及森林资源二类调查数据的多源数据估算毛竹林面积和郁闭度,CatBoost算法具有最佳的性能指标,其估算结果可为竹林资源的监测提供重要借鉴。
Abstract:
Based on images from Sentinel-2 medium resolution remote sensing imagery, digital elevation model (DEM) and bamboo (Phyllostachys edulis) data from forest management survey of Anji county, Zhejiang province in 2018, estimation was made on area and crown density of bamboo forest by CatBoost, random forest (RF) and support vector machine (SVM) model, dichotomy idea and multi-classification. The result demonstrated that classification accuracy of bamboo forest area was 97.75% by CatBoost, 96.54% by RF and 95.05% by SVM, the kappa coefficients were 0.96, 0.93 and 0.9 respectively. The result of Moso bamboo forest canopy density estimated were as follows: the overall classification accuracy of the above three algorithms were 73.38%, 73.49% and 69.9% respectively, and the kappa coefficients were 0.52, 0.52 and 0.45 respectively. The estimation accuracy of models was greatly improved after the samples divided into two categories (large sample and small sample) and modeled separately. The estimation accuracy of crown density of each model was greater than 78%, especially by CatBoost, 83.45%.

参考文献/References:

[1] MANANDHAR R,KIM J H,KIM J T. Environmental, social and economic sustainability of bamboo and bamboo-based construction materials in buildings[J]. J Asian Arch Build Engin,2019,40(24):9541-9555.
[2] 魏芬娟,陈崇成,叶晓燕,等. 森林资源调查与督查的移动GIS研发及应用[J]. 计算机工程与应用,2020,56(6):231-238.
[3] 蒲梦馨,倪忠云,赵银兵,等. 竹林信息遥感提取方法研究进展[J]. 四川林业科技,2021,42(4):122?131.
[4] 高国龙,杜华强,韩凝,等. 基于特征优选的面向对象毛竹林分布信息提取[J]. 林业科学,2016,52(9):77-85.
[5] 吴梓尚,林辉,孙华,等. 基于MODIS影像的森林类型决策树分类方法研究[J]. 中南林业科技大学学报,2017,37(9):80-85.
[6] 杜华强,孙晓艳,韩凝,等. 综合面向对象与决策树的毛竹林调查因子及碳储量遥感估算[J]. 应用生态学报,2017,28(10):3163-3173.
[7] 崔璐,杜华强,周国模,等. 决策树结合混合像元分解的中国竹林遥感信息提取[J]. 遥感学报,2019,23(1):166-176.
[8] 熊恬苇. 江西省竹林时空变化遥感研究[D]. 南昌:江西师范大学,2018.
[9] 严欣荣,张美曼,郑亚雄,等. 基于Sentinel-2的丛生竹林信息提取方法比较及分布特征[J]. 生态学杂志,2020,39(3):1056-1066.
[10] 徐辉,潘萍,杨武,等. 基于多源遥感影像的森林资源分类及精度评价[J]. 江西农业大学学报,2019,41(4):751-760.
[11] 钟滨,廖永皓,蔡海生. 基于光谱与纹理特征的高分二号竹林信息提取以庐山自然保护区为例[J]. 测绘与空间地理信息,2020,43(12):8-13.
[12] 王聪. 基于多源遥感数据的毛竹林冠层郁闭度多尺度反演研究[D]. 杭州:浙江农林大学,2015.
[13] DU H Q,MAO F J,LI X J,et al. Mapping global bamboo forest distribution using multisource remote sensing data[J]. IEEE J SelecTop App Earth Observ Remot Sens,2018,11(5):1-14.
[14] DIDA J J V,Araza A B,Eduarte G T,et al. Towards nationwide mapping of bamboo resources in the Philippines: testing the pixel-based and fractional cover approaches[J]. Int J Remot Sens,2021,42(9):3380-3404.
[15] WANG Y,JIN J X,JIANG H,et al. Satellite-based detection of bamboo expansion over the past 30 years in Mount Tianmushan, China[J]. Int J Remot Sens,2016,37(13-14):2908-2922.
[16] WATANABE S,SUMI K,ISE T. Identifying the vegetation type in Google Earth images using a convolutional neural network: a case study for Japanese bamboo forests[J]. BMC Ecol,2020,20(1):65-79
[17] 张瑞英,庞勇,李增元,等. 结合机载LiDAR和LANDSAT ETM+数据的温带森林郁闭度估测[J]. 植物生态学报,2016,40(2):102-115.
[18] 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),2016,59(05):99-106.
[19] 杜华强,周国模,葛宏立,等. 基于TM数据提取竹林遥感信息的方法[J]. 东北林业大学学报,2008,36(3):35-38.
[20] 蔡林菲,吴达胜,方陆明,等. 基于XGBoost的高分二号影像树种识别[J]. 林业资源管理,2019,0(5):44-51.
[21] 李倩楠. 基于不同算法的DEM地面曲率提取的比较分析[J]. 首都师范大学学报(自然科学版),2016,37(5):82-85.
[22] RICH P M,DUBAYAH R,HETRICK W A,et al. Using viewshed models to calculate intercepted solar radiation: applications in ecology[C]. Proceedings of American society for photogrammetry and remote sensing technical. Bethesda:American Society for Photogrammetry and Remote Sensing,1994:524-529.
[23] FU P,RICH P M. A geometric solar radiation model with applications in landscape ecology[D]. Lawrence Kansas:University of Kansas,2000.
[24] QUINN P,BEVEN K J,LAMB R. The ln (α/tanβ) index: how to calculate it and how to use it within the TOPMOD-EL framework[J]. Hydrol Proc,2010,9(2):161-182.
[25] 骆爱琴,曾玲梅,王勤. 安徽铜陵毛竹林生长与土壤特点分析[J]. 山东林业科技,2010,40(2):40-41.
[26] IBRAHIM A A,RIDWAN R L,MUHAMMED M M,et al. Comparison of the catBoost classifier with other machine learning methods[J]. Int J Adv Comput Sci App,2020,11(11):11.
[27] 董师师,黄哲学. 随机森林理论浅析[J]. 集成技术,2013,2(1):1-7.
[28] 张浩然,韩正之,李昌刚. 支持向量机[J]. 计算机科学,2002,29(12):135-137,142.
[29] THOMAS J,JOUVE P E,NICOLOYANNIS N. Optimisation and evaluation of random forests for imbalanced datasets[J]. Lect Note Artif Intell, 2006,622-631.

相似文献/References:

[1]宋思睿,烟亚萍,宣丹娟,等.集约经营毛竹林土壤酶活性对模拟氮沉降的初期响应[J].浙江林业科技,2015,35(02):62.
[2]唐隆校,周智峰,潘建华.林地覆盖对毛竹鞭笋生长的影响[J].浙江林业科技,2015,35(04):81.
 [J].Journal of Zhejiang Forestry Science and Technology,2015,35(03):81.
[3]钟子龙,周紫球,陆媛媛,等.维持覆盖毛竹林地下竹鞭正常生长的技术研究[J].浙江林业科技,2015,35(04):85.
 [J].Journal of Zhejiang Forestry Science and Technology,2015,35(03):85.
[4]王利平,朱炜,沈泉,等.毛竹林下套种三叶崖爬藤的研究[J].浙江林业科技,2016,36(03):73.
 WANG Li-ping,ZHU Wei,SHEN Quan,et al.Experiment on Interplanting Tetrastigma hemsleyanum under Phyllostachys heterocycla cv. pubescens Stand[J].Journal of Zhejiang Forestry Science and Technology,2016,36(03):73.
[5]张飞英,刘亚群,徐瑞英,等.长期施肥下毛竹林土壤的肥力质量研究[J].浙江林业科技,2016,36(04):19.
 ZHANG Fei-ying,LIU Ya-qun,XU Rui-ying,et al.Study on Soil Quality under Phyllostachys heterocycla cv. pubescensForest with Fertilization[J].Journal of Zhejiang Forestry Science and Technology,2016,36(03):19.
[6]张飞英,刘亚群,柏明娥.长期不同施肥方式对毛竹林土壤酸化过程的影响[J].浙江林业科技,2017,37(02):60.[doi:10.3969/j.issn.1001-3776.2017.02.009]
 ZHANG Fei-ying,LIU Ya-qun,BAI Ming-e.Effects of Different Fertilizations on Soil Acidification of Phyllostachys heterocycla cv.Pubescens Stands[J].Journal of Zhejiang Forestry Science and Technology,2017,37(03):60.[doi:10.3969/j.issn.1001-3776.2017.02.009]
[7]陆超,沈建军,王秀云,等.温州城区绿地生物量时空动态变化研究[J].浙江林业科技,2017,37(04):18.[doi:10.3969/j.issn.1001-3776.2017.04.004]
 LU Chao,SHEN Jian-jun,WANG Xiu-yun,et al.Spatiotemporal Dynamics of Biomass of Green Space in the City Proper of Wenzhou[J].Journal of Zhejiang Forestry Science and Technology,2017,37(03):18.[doi:10.3969/j.issn.1001-3776.2017.04.004]
[8]毛朝明,蒋灵华.松阳县毛竹林地土壤养分空间变异特征分析[J].浙江林业科技,2020,40(02):65.[doi:10.3969/j.issn.1001-3776.2020.02.010]
 MAO Zhao-ming,JIANG Ling-hua.Spatial Variability of Soil Nutrient Content in Phyllostachys edulis Stands in Songyang County[J].Journal of Zhejiang Forestry Science and Technology,2020,40(03):65.[doi:10.3969/j.issn.1001-3776.2020.02.010]
[9]张 乐,王志辉,徐惠军,等.基于包络线去除法的森林树种及树种组分类[J].浙江林业科技,2020,40(02):91.[doi:10.3969/j.issn.1001-3776.2020.02.014]
 ZHANG Le,WANG Zhi-hui,XU Hui-jun,et al.Classification of Tree Species and Groups based on Envelope Removal[J].Journal of Zhejiang Forestry Science and Technology,2020,40(03):91.[doi:10.3969/j.issn.1001-3776.2020.02.014]

备注/Memo

备注/Memo:
收稿日期:2022-12-30;修回日期:2023-03-05
基金项目:浙江省科技重点研发计划资助项目(2018C02013)
作者简介:王广科,硕士,从事农业工程与信息技术研究;E-mail:876294612@qq.com。
更新日期/Last Update: 2023-05-20