Please wait a minute...
img

官方微信

遥感技术与应用  2019, Vol. 34 Issue (5): 959-969    DOI: 10.11873/j.issn.1004-0323.2019.5.0959
林业遥感专栏     
基于KNN-FIFS的内蒙古根河森林郁闭度遥感估测研究
孙珊珊1(),田昕1(),谷成燕2,韩宗涛1,王崇阳1,张兆鹏3
1. 中国林业科学研究院资源信息研究所 北京100091
2. 国家林业和草原管理局林业产品工业规划设计研究院 北京 100010
3. 自然资源部第一大地测量队 陕西 西安 710054
Estimation of Forest Canopy Closure by the KNN-FIFS Method in the Genhe of Inner Mongolia
Shanshan Sun1(),Xin Tian1(),Chengyan Gu2,Zongtao Han1,Chongyang Wang1,Zhaopeng Zhang3
1. Institute of Forest Resource Information Techniques, Chinese Academy of Forestry, Beijing 100091, China
2. Planning and Design Institute of Forestry Product Industry, National Forestry and Grassland Administrition, Beijing 100010, China
3. The First Geodetic Surveying Brigade of MNR, Xi'an 710054, China
 全文: PDF(4036 KB)   HTML
摘要:

为探索国产高分一号宽幅(GF-1 Wide Field of View,GF-1 WFV)数据以及具有宽覆盖、红边波段(Red-Edge band,RE)的高分六号(GF-6)卫星数据在森林郁闭度(Forest Canopy Closure,FCC)定量反演中的潜力,本研究以GF-1 WFV多光谱数据为基础,添加哨兵2号(Sentinel-2A)红边波段,模拟GF-6红边波段特性,并提取相关纹理信息(Texture Information,TI)、植被指数(Vegetation Index,VI)和红边指数(Red- edge Index,RI),同时添加太阳入射角的余弦值cosi和1/cosi进一步探究了地形因素(Topographic Factors,TF)对FCC估测的影响,利用快速迭代特征选择的k-NN(k-Nearest Neighbor with Fast Iterative Features Selection,KNN-FIFS)模型,实现了内蒙古大兴安岭根河研究区FCC的定量反演,并对比逐步多元线性回归(Stepwise Multiple Linear Regressions,SMLR)和支持向量机(Support Vector Machine,SVM)估测结果。通过44块调查样地实测数据验证发现:基于GF-1 WFV估测的FCC与实测数据具有很好的一致性,R2=0.52,RMSE=0.08;GF-1 WFV+VI+TI估测结果为R2=0.56,RMSE=0.08;GF-1 WFV+RE+RI+TI的精度明显提高,R2=0.63,RMSE=0.07;GF-1 WFV+RE+RI+TI+TF的精度最高,R2=0.68,RMSE=0.07,并高于SMLR(R2=0.39,RMSE=0.10)和SVM(R2=0.49,RMSE=0.10)方法。KNN-FIFS方法比SMLR和SVM方法更适用于FCC遥感估测,且添加红边信息经地形校正后,能有效提高FCC的估测精度。

关键词: 森林郁闭度高分一号高分六号红边KNN-FIFS根河森林    
Abstract:

Aiming at exploring the potentials of Gaofen-1 (GF-1) WFV data and Gaogen-6 (GF-6) satellite data in quantitative inversion of Forest Canopy Closure (FCC), based on GF-1data, the simulated GF-6 data by adding two Sentinel-2A red-edge bands (RE) into GF-1 WFV multispectral data and the extracted relevant Texture Information (TI) ,Vegetation Index (VI) and Red edge Index (RI) , a k- Nearest Neighbor with Fast Iterative Features Selection( KNN-FIFS) method, was used to estimate the Forest Canopy Closure (FCC) in the Genhe of the Great Khingan, Inner Mongolia. Besides that, the impact of terrain was further explored by adding Topographic Factors (TF) into the feature compositions. The verification using 44 field samples and the Leave-One-Out (LOO) method showed that: FCC estimation based on GF-1 WFV is in good agreement with measured data, with R2 = 0.52, RMSE = 0.08; the GF-1 WFV+VI+TI’s has R2 = 0.56, RMSE = 0.08; the GF-1 WFV+RE+RI+TI’s has been significantly improved with R2=0.63 and RMSE=0.07; and highest accuracy from the GF-1 WFV+RE+RI+TI+TF composition with R2=0.68 and RMSE=0.07 was superior to the results from both stepwise multiple linear regressions (SMLR) (R2=0.39, RMSE=0.10) and support vector machine (SVM) (R2=0.49, RMSE=0.10) methods. It indicated that the KNN-FIFS method is more reliable for FCC estimation than both SMLR and SVM methods, and the simulated GF-6 data with red-edge information can effectively improve the estimation accuracy of FCC, especially after topographic correction.

Key words: Forest canopy closure    GF-1 WFV    GF-6    Red-edge    KNN-FIFS    Forest of Genhe
收稿日期: 2019-02-21 出版日期: 2019-12-05
ZTFLH:  TP79  
基金资助: 中央级公益性科研院所基金青年人才项目“森林资源动态变化时空连续监测方法研究”(CAFYBB2017QC005);国家自然科学基金项目“森林地上生物量动态信息时空协同分析及建模”(41871279)
通讯作者: 田昕     E-mail: shanss_caf@163.com;tianxin@ifrit.ac.cn
作者简介: 孙珊珊(1994-),女,山东青岛人,硕士研究生,主要从事遥感技术应用研究。E?mail:shanss_caf@163.com
服务  
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章  
孙珊珊
田昕
谷成燕
韩宗涛
王崇阳
张兆鹏

引用本文:

孙珊珊,田昕,谷成燕,韩宗涛,王崇阳,张兆鹏. 基于KNN-FIFS的内蒙古根河森林郁闭度遥感估测研究[J]. 遥感技术与应用, 2019, 34(5): 959-969.

Shanshan Sun,Xin Tian,Chengyan Gu,Zongtao Han,Chongyang Wang,Zhaopeng Zhang. Estimation of Forest Canopy Closure by the KNN-FIFS Method in the Genhe of Inner Mongolia. Remote Sensing Technology and Application, 2019, 34(5): 959-969.

链接本文:

http://www.rsta.ac.cn/CN/10.11873/j.issn.1004-0323.2019.5.0959        http://www.rsta.ac.cn/CN/Y2019/V34/I5/959

图1  研究区地理位置及野外调查样地分布(图中影像为GF-1 WFV,R:band 3,G:band 2,B:band 1)
图2  研究区高程图
图3  研究区cosi值分布图
名称简称说明
均值(Mean)Mi

i=1,2,…6,分别对应

GF-1WFV数据中的

4个多光谱波段和哨兵

2A数据的B5和B6红

边波段

方差(Variance)Vi
均一性(Homogeneity)Hi
对比度(Contrast)Coi
相异性(Dissimilarity)Di
熵(Entropy)Ei
二阶矩(Second moment)Si
相关性(Correlation)Cri
表1  遥感数据纹理信息
序号波段组合
1GF-1 WFV
2GF-1 WFV、VI、TI
3GF-1 WFV、RE、RI、TI
4GF-1 WFV、RE、RI、TI、TF
表2  KNN-FIFS实验不同波段组合
图4  KNN-FIFS算法流程图
图5  不同特征组合估测FCC交叉验证结果
图6  估测FCC交叉验证结果
图7  研究区FCC分布图
图8  研究区FCC分级图
图9  光谱响应曲线
排序k值优选特征RMSE
13M2、M4、S3、DVIr、cosi0.067 260
23M2、M4、S3、DVIr、Cr20.067 262
33M2、M4、S3、DVIr0.069 961
45E2、NDRE、S3、V1、cosi0.075 287
55E2、NDRE、S3、V10.076 796
65M5、B、M4、S3、V10.077 759
74E4、G、S2、NDRE、SRr、Cr20.078 086
表3  KNN-FIFS实验运算结果
图10  研究区样地坡度值
1 Jennings S B, Brown N D, Sheil D. Assessing Forest Canopies and Understorey Illumination: Canopy Closure, Canopy Cover and Other Measures[J]. Forestry, 1999, 72(1):59-74.
2 Penman J, Gytarsky M, Hiraishi T, et al. Good Practice Guidance for Land Use, Land- Use Change and Forestry, Intergovernmental Panel on Climate Change: IPCC National Greenhouse Gas Inventories Program[C]∥ Institute for Global Environmental Strategies (IGES) for the IPCC, Kanagawa Japan, 2003.
3 Li Chonggui, Cai Tijiu.Effect of Forest Canopy Density on Stock Volume Estimation [J]. Journal of Northeast Forestry University,2006,34(1):15-17.李崇贵, 蔡体久. 森林郁闭度对蓄积量估测的影响规律[J].东北林业大学学报,2006,34(1):15-17.
4 Alike S, Samanta S. Land Use/Land Cover and Forest Canopy Density Monitoring of Wafi-Golpu Project Area, Papua New Guinea[J]. Journal of Geoscience and Environment Protection, 2016, 4: 1-14.
5 Li Yuhao, Shi Tian.Estimating the Canopy Density through Color Features of Aerial Digital Images[J]. Journal of Northwest Forestry University, 2014,29(1):148-154.李宇昊, 石 田. 利用航空数码影像的色彩特征估测森林郁闭度[J]. 西北林学院学报,2014,29(1):148-154.
6 Wang Qiang, SiQinbilige, Xiao Haitao, et al. Prospect and Discussion on Measuring Methods of Forest Canopy Density [J]. Inner Mongolia Forestry Investigation and Design, 2014, 37(1):82-83.王强, 斯琴毕力格, 肖海涛,等.森林郁闭度测定方法探讨与展望[J].内蒙古林业调查设计, 2014, 37(1):82-83.
7 Wang Cong, Du Huaqiang, Zhou Guomo, et al. Retrieval of Crown Closure of Moso Bamboo Forest Using Unmanned Aerial Vehicle(UAV) Remotely Sensed Imagery based on Geometric-optical Model[J]. Chinese Journal of Applied Ecology, 2015, 26(5):1501-1509.王聪, 杜华强, 周国模,等. 基于几何光学模型的毛竹林郁闭度无人机遥感定量反演[J]. 应用生态学报, 2015, 26(5):1501-1509.
8 Tian X, Su Z, Chen E, et al. Estimation of Forest above-ground Biomass Using Multi-parameter Remote Sensing Data over a Cold and Arid Area[J].International Journal of Applied Earth Observation and Geoinformation,2012,14(1): 160-168.
9 Qi Yujiao, Li Fengri.Remote Sensing Estimation of Aboveground Forest Carbon Storage in Daxing’an Mountains based on KNN Method [J]. Scientia Silvae Sinicae, 2015,51(5): 46-55.戚玉娇, 李凤日. 基于KNN方法的大兴安岭地区森林地上碳储量遥感估算[J]. 林业科学, 2015,51(5): 46-55.
10 Zhang Jin, Li Xiaosong, Wu Bingfang.Forest Cover Estimation based on Classification and Regression Trees of Miyun Reservoir Upstream Area [J]. Remote Sensing Technology and Application, 2014, 29(3):394-400.张瑾, 李晓松, 吴炳方. 基于分类回归树的密云水库上游森林覆盖度遥感估算[J]. 遥感技术与应用, 2014, 29(3):394-400.
11 Franco-Lopez H, Ek A R, Bauer M E. Estimation and Mapping of Forest Stand Density, Volume, and Cover Type Using the k-nearest Neighbors Method[J]. Remote Sensing of Environment, 2001, 77(3): 251-274.
12 Crookston N L, Finley A O. yaImpute: An R Package for kNN Imputation [J]. Journal of Statistical Software, 2008, 23(16):1-16.
13 Troyanskaya O, Cantor M, Sherlock G, et al. Missing Value Estimation Methods for DNA Microarrays[J]. Bioinformatics, 2001, 17(6): 520-525.
14 Tian X, Yan M, Christiaan V D T, et al. Modeling Forest above-ground Biomass Dynamics Using Multi-source Data and Incorporated Models: A Case Study over the Qilian Mountains[J]. Agricultural and Forest Meteorology, 2017, 246:1-14.
15 Wilson B T, Lister A J, Riemann R I. A Nearest-neighbor Imputation Approach to Mapping Tree Species over Large Areas Using Forest Inventory Plots and Moderate Resolution Raster Data[J]. Forest Ecology and Management, 2012, 271: 182-198.
16 Reese H, Nilsson M, Sandström P, et al. Applications Using Estimates of Forest Parameters Derived from Satellite and Forest Inventory Data[J]. Computers and Electronics in Agriculture, 2002, 37(1): 37-55.
17 Tian X, Li Z, Su Z, et al. Estimating Montane Forest above-ground Biomass in the Upper Reaches of the Heihe River Basin Using Landsat-TM Data[J]. International Journal of Remote Sensing, 2014, 35(21): 7339-7362.
18 Han Zongtao, Jiang Hong, Wang Wei, et al. Forest above-ground Biomass Estimation Using KNN-FIFS Method based on Multi-Source Remote Sensing Data[J]. Scientia Silvae Sinicae,2018, 54(9):70-79.韩宗涛, 江洪, 王威,等. 基于多源遥感的森林地上生物量KNN-FIFS估测[J]. 林业科学, 2018, 54(9):70-79.
19 Breiman L. 1996. Heuristics of Instability and Stabilization in Model Selection[J]. The Annals of Statistics, 24(6): 2350-2383.
20 Zhang Ruiying, Pang Yong, Li Zengyuan, et al. Canopy Closure Estimation in a Temperate Forest Using Airborne LiDAR and Landsat ETM+ Data [J]. Acta Phytoecologica Sinica, 2016, 40(2):102-115.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.
21 Xu Ding, Peng Daoli.Estimation of Forest Canopy Closure based on Dimidiate Pixel Model [J]. Journal of Northeast Forestry University, 2013,41(2):119-122.徐 定, 彭道黎. 基于像元二分模型的森林郁闭度估测方法[J]. 东北林业大学学报,2013,41(2):119-122.
22 Kahriman A, Günlü A, Karahalil U. Estimation of Crown Closure and Tree Density Using Landsat TM Satellite Images in Mixed Forest Stands[J]. Journal of the Indian Society of Remote Sensing, 2014, 42(3):559-567.
23 Yan Min, Li Zengyuan, Chen Erxue, et al. Vegetation Fractional Coverage Change in Daxinganling Genhe Forest Reserve of Inner Mongolia [J]. Chinese Journal of Ecology, 2016, 35 (2): 508-515.闫敏, 李增元, 陈尔学,等. 内蒙古大兴安岭根河森林保护区植被覆盖度变化[J]. 生态学杂志, 2016, 35(2):508-515.
24 Jia K, Liang S, Gu X, et al. Fractional Vegetation Cover Estimation Algorithm for Chinese GF-1 Wide Field View Data [J]. Remote Sensing of Environment, 2016, 177:184-191.
25 Viña A, Gitelson A A, Nguy-Robertson, et al. Comparison of Different Vegetation Indices for the Remote Assessment of Green Leaf Area Index of Crops[J]. Remote Sensing of Environment, 2011, 115(12): 3468-3478
26 Nguy-Robertson A, Gitelson A, Peng Y, et al. Green Leaf area Index Estimation in Maize and Soybean: Combining Vegetation Indices to Achieve Maximal Sensitivity[J]. Agronomy Journal, 2012, 104(5): 1336-1347
27 Yan Wei, Zhou Wen, Yi Lilong, et al. Research Progress of Remote Sensing Classification and Change Monitoring on Forest Types[J]. Remote Sensing Technology and Application, 2019, 34(3): 445-454.
27 颜伟, 周雯, 易利龙, 等. 森林类型遥感分类及变化监测研究进展[J]. 遥感技术与应用, 2019, 34(3): 445-454.
28 Ren Chong. Forest Types Precise Classification and Forest Resources Change Monitoring based on Medium and High Spatial Resolution Remote Sensing Images [D]. Beijing:Chinese Academy of Forestry,2016.任冲. 中高分辨率遥感影像森林类型精细分类与森林资源变化监测技术研究[D]. 北京:中国林业科学研究院, 2016.
29 Dong Dejin, Zhou Guomo, Du Huaqiang, et al. Effects of Topographic Correction with Six Correction Models on Phyllostachys Praecox Forest Aboveground Biomass Estimation[J]. Scientia Silvae Sinicae, 2011, 47(12):1-8.董德进, 周国模, 杜华强,等. 6种地形校正方法对雷竹林地上生物量遥感估算的影响[J]. 林业科学, 2011, 47(12):1-8.
30 Yao Chen, Huang Wei, Li Xianhua. Evaluation of Topographical Influence on Vegetation Indices of Rugged Terrain [J]. Remote Sensing Technology and Application, 2009,24(4):496- 501.姚晨, 黄微, 李先华. 地形复杂区域的典型植被指数评估[J]. 遥感技术与应用,2009,24(4):496-501.
31 Anderson G P, Felde G W, Hoke M L, et al. MODTRAN4-based Atmospheric Correction Algorithm: FLAASH (Fast Line-of-sight Atmospheric Analysis of Spectral Hypercubes)[J]∥ AeroSense 2002. International Society for Optics and Photonics, 4725:65-71.
32 Rouse J W J, Haas R H, Schell J A, et al. Monitoring Vegetation Systems in the Great Plains with ERTS[R]. NASA Special Publication,1973, 351: 309.
33 Birth G S, Mcvey G R. Measuring the Color of Growing Turf with a Reflectance Spectrophotometer [J]. Agronomy Journal, 1968, 60(6):640-643.
34 Tucker C J. Red and Photographic Infrared Linear Combinations for Monitoring Vegetation[J]. Remote Sensing of Environment, 1979, 8(2):127-150.
35 Gitelson A A, Keydan G P, Merzlyak M N. Three‐band Model for Noninvasive Estimation of Chlorophyll, Carotenoids, and Anthocyanin Contents in Higher Plant Leaves[J]. Geophysical Research Letters, 2006, 33(11):431-433.
36 Haralick R M, Shanmugam K. Textural Features for Image Classification[J]. IEEE Transactions on Systems, Man, and Cybernetics, 1973, (6): 610-621.
37 Tokola T, Pitkänen J, Partinen S, et al. Point Accuracy of a Non-parametric Method in Estimation of Forest Characteristics with Different Satellite Materials[J]. International Journal of Remote Sensing, 1996,17(12): 2333-2351.
38 Cortes C, Vapnik V. Support-vector Networks [J]. Machine Learning, 1995, 20(3): 273-297.
39 Burges C J C. A Tutorial on Support Vector Machines for Pattern Recognition [J]. Data Mining and Knowledge Discovery, 1998, 2(2): 121-167.
40 Chang C C, Lin C J. LIBSVM: A Library for Support Vector Machines [J]. ACM Transactions on Intelligent Systems and Technology (TIST), 2011, 2(3): 27.
41 Lawrence R L, Ripple W J. Comparisons among Vegetation Indices and Bandwise Regression in a Highly Disturbed, Heterogeneous Landscape: Mount St. Helens, Washington [J]. Remote Sensing of Environment, 1998, 64(1):91-102.
42 Foody G M, Cutler M E, Mcmorrow J, et al. Mapping the Biomass of Bornean Tropical Rain Forest from Remotely Sensed Data[J]. Global Ecology & Biogeography, 2001, 10(4):379-387.
43 Næsset, Erik, Gobakken T. Estimation of above- and Below-ground Biomass Across Regions of the Boreal Forest Zone Using Airborne Laser[J]. Remote Sensing of Environment, 2008, 112(6):3079-3090.
44 Filella I, Peñuelas J. The Red Edge Position and Shape as Indictors of Plant Chlorophyll Content, Biomass and Hydric Status[J]. International Journal of Remote Sensing,1994,15(7):1459-1470.
45 Hansen P M, Schjoerring J K. Reflectance Measurement of Canopy Biomass and Nitrogen Statue in Wheat Crops Using Normalized Difference Vegetation Indices and Partial Least Squares Regression[J]. Remote Sensing of Environment,2003,86(4):542-553.
46 Holben B N, Justice C O. The Topographic Effect on Spectral Response from Nadir-pointing Sensors [J]. 1980,46(9):1191-1200.
47 Zhu Gaolong, Liu Yibo, Ju Weimin, et al. Evaluation of Topographic Effects on Four Commonly Used Vegetation Indices[J]. Journal of Remote Sensing, 2013,17(1):210-234.
47 朱高龙,柳艺博,居为民,等. 4 种常用植被指数的地形效应评估[J].遥感学报,2013,17(1):210-234.
[1] 孙红, 田昕, 闫敏, 李增元, 陈尔学, 孙珊珊, 王崇阳. 内蒙古大兴安岭根河植被覆盖度动态变化及影响因素的分析[J]. 遥感技术与应用, 2018, 33(6): 1159-1169.
[2] 江东,陈帅,丁方宇,付晶莹,郝蒙蒙. 基于面向对象的遥感影像分类研究——以河北省柏乡县为例[J]. 遥感技术与应用, 2018, 33(1): 143-150.
[3] 于博文,田淑芳. “高分一号”卫星数据几何校正研究[J]. 遥感技术与应用, 2017, 32(1): 133-139.
[4] 张胜敏,周美玲,司一丹,王中挺. 深蓝算法应用于GF-1 16 m相机反演陆地气溶胶[J]. 遥感技术与应用, 2016, 31(4): 709-713.
[5] 曹学诚,刘周周. 基于PostGIS的高分一号卫星影像管理与Web应用研究[J]. 遥感技术与应用, 2016, 31(2): 285-289.
[6] 刘锟,付晶莹,李飞. 高分一号卫星4种融合方法评价[J]. 遥感技术与应用, 2015, 30(5): 980-986.
[7] 梁文秀,李俊生,周德民,申茜,张方方. 面向内陆水环境监测的GF-1卫星WFV数据特征评价[J]. 遥感技术与应用, 2015, 30(4): 810-818.
[8] 邬登巍,吴昀昭,马宏瑞 . 植物污染胁迫遥感监测研究综述[J]. 遥感技术与应用, 2009, 24(2): 238-245.
[9] 卢艳丽, 李少昆, 白由路, 谢瑞芝, 宫永梅. 冬小麦冠层光谱红边参数的变化及其与氮素含量的相关分析[J]. 遥感技术与应用, 2007, 22(1): 1-7.
[10] 谭昌伟,王纪华,黄文江,刘良云,赵春江,郭新宇. 不同株型夏玉米群体冠层反射光谱特征及其应用研究[J]. 遥感技术与应用, 2004, 19(4): 225-231.
[11] 黄文江,王纪华,刘良云,赵春江,王锦地,杜小鸿. 冬小麦红边参数变化规律及其营养诊断[J]. 遥感技术与应用, 2003, 18(4): 206-211.
[12] 金震宇,田庆久,惠凤鸣,陆建飞. 水稻叶绿素浓度与光谱反射率关系研究[J]. 遥感技术与应用, 2003, 18(3): 134-137.