基于多源遥感的森林地上生物量KNN-FIFS估测

doi:10.11707/j.1001-7488.20180909

摘要/Abstract

摘要： [目的]针对多源遥感数据及其派生特征因子数据维度高、信息冗余、易造成估测模型过拟合等问题，从高维度遥感特征因子中高效优化特征组合，优化区域森林地上生物量（AGB）的k最近邻（k-NN）估测模型。[方法]提出基于快速迭代特征选择的k最近邻法（KNN-FIFS），以森林资源样地调查数据计算的森林AGB为参考，以留一法交叉验证（LOO）相应的k-NN模型反演的森林AGB均方根误差（RMSE）最小为原则，依次迭代选取遥感特征，优化区域森林AGB的k-NN估测模型。以大兴安岭根河森林保护区为研究区，结合Landsat-8 OLI各波段光谱信息、植被指数、纹理、地形因子、机载合成孔径雷达（SAR） P-波段HV极化后向散射强度信息（P_HV）以及森林资源样地调查数据，利用KNN-FIFS方法估测研究区森林AGB，并与多元线性逐步回归法（SMLR）进行对比分析。[结果]利用KNN-FIFS方法，得到当k为3，特征组合为P_HV、短波红外波段一均一性（H6）、短波红外波段一二阶矩（S6）、短波红外波段二二阶矩（S7）、海蓝波段相关性（Cr1）、近红外波段相关性（Cr5）、海蓝波段相异性（D1）、增强型植被指数（EVI）时，研究区森林AGB估测结果最优，其精度（R²=0.77，RMSE=22.74 t·hm^-2）显著优于SMLR估测精度（R²=0.53，RMSE=32.37 t·hm^-2）。[结论]KNN-FIFS方法相比SMLR更适用于森林AGB多源遥感估测；KNN-FIFS方法可以从高维度遥感特征因子中高效选取相关特征进行森林AGB估测。

关键词: KNN-FIFS, 特征选择, 地上生物量

Abstract: [Objective] Aiming at the over-fitting problem caused by information redundancy from multi-source remote sensing data and their derived high-dimensional features, this study is to effectively pre-select the optimal feature combination to optimize the k-nearest neighbor (k-NN) for regional forest above-ground biomass (AGB) estimation.[Method] This study proposed a fast iterative features selection method for k-NN method (KNN-FIFS). This method iteratively pre-select the optimal features which determined by the minimum root mean square error (RMSE) between the measured forest AGB values and the k-NN estimates based on the leave-one-out (LOO) cross-validation. Based on KNN-FIFS, multi-source data, including Landsat-8 OLI and its vegetation indices, texture metrics, topographic factors, HV polarization of P-band synthetic aperture radar (SAR) data, and forest inventory data (P_HV), were used to estimate forest AGB over Daxing'an Mountain Genhe forest reserve located in Inner Mongolia. Afterwards, the model behaviors between KNN-FIFS and stepwise multiple linear regression (SMLR) method were compared.[Result] For KNN-FIFS method, the best configuration was that one with k of 3, the remotely sensed features using P_HV, second moment of 1^st and 2^nd short-wave infrared bands (S6,S7), homogeneity of 1^st short-wave infrared band (H6), correlation of coastal aerosol (Cr1), correlation of the near infrared (Cr5), dissimilarity of coastal aerosol (D1) and the enhanced vegetation index (EVI). This configuration generated the most accurate estimates with R²=0.77 and RMSE=22.74 t·hm^-2,which performed much better than SMLR with R²=0.53 and RMSE=32.37 t·hm^-2.[Conclusion] KNN-FIFS is a more suitable method for forest AGB estimation than SMLR. KNN-FIFS can efficiently select the optimal feature combination to estimate regional forest AGB by use of multi-source remote sensing data with high-dimensional information.

Key words: KNN-FIFS, feature selection, above-ground biomass(AGB)

中图分类号:

S757

韩宗涛, 江洪, 王威, 李增元, 陈尔学, 闫敏, 田昕. 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.

Han Zongtao, Jiang Hong, Wang Wei, Li Zengyuan, Chen Erxue, Yan Min, Tian Xin. 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.

参考文献

曹庆先, 徐大平, 鞠洪波. 2011. 基于TM影像纹理与光谱特征和KNN方法估算5种红树林群落生物量. 林业科学研究, 24(2): 144-150.
(Cao Q X, Xu D P, Ju H B. 2011. Biomass estimation of five kinds of mangrove community with the KNN method based on the spectral information and textural features of TM images. Forest Research, 24(2): 144-150. [in Chinese])
陈传国,朱俊凤. 1989. 东北主要林木生物量手册. 北京:中国林业出版社.
(Chen C G, Zhu J F. 1989. Woody biomass manual of typical species in the northeast of China. Beijng: China Forestry Publishing House. [in Chinese])
冯琦, 陈尔学, 李增元, 等. 2016. 基于机载P-波段全极化SAR数据的复杂地形森林地上生物量估测方法. 林业科学, 52(3): 10-21.
(Feng Q, Chen E X, Li Z Y, et al. 2016. Forest Above-ground biomass estimation method for rugged terrain based on airborne P-band PolSAR data. Scientia Silvae Sinicae, 52(3): 10-21. [in Chinese])
郭云, 李增元, 陈尔学, 等. 2015. 甘肃黑河流域上游森林地上生物量的多光谱遥感估测. 林业科学, 51(1): 140-149.
(Guo Y, Li Z Y, Chen E X, et al. 2015. Estimating forest above-ground biomass in the upper reaches of Heihe River basin using multi-spectral remote sensing. Scientia Silvae Sinicae, 51(1): 140-149. [in Chinese])
国庆喜, 张锋. 2003. 基于遥感信息估测森林的生物量. 东北林业大学学报, 31(2): 13-16.
(Guo Q X, Zhang F. 2003. Estimation of forest biomass based on remote sensing. Journal of Northeast Forestry University, 31(2): 13-16. [in Chinese])
郭志华, 彭少麟, 王伯荪. 2002. 利用TM数据提取粤西地区的森林生物量. 生态学报, 22(11): 1832-1839.
(Guo Z H, Peng S L, Wang B S. Estimating forest biomass in western Guangdong using Landsat TM data. Acta Ecologica Sinica, 22(11): 1832-1839. [in Chinese])
胡凯龙, 刘清旺, 穆喜云. 2015. 差分 GNSS 系统在大兴安岭地区森林资源调查中的精度分析. 林业调查规划, 40(4): 1-6.
(Hu K L, Liu Q W, Mu X Y. 2015. Differential GNSS application on location precision analysis of forest resource investigation in Daxinganling region. Forest Inventory and Planning, 40(4): 1-6. [in Chinese])
李春梅, 张王菲, 李增元, 等. 2016. 基于多源数据的根河实验区生物量反演研究. 北京林业大学学报, 38(3): 64-72.
(Li C M, Zhang W F, Li Z Y, et al. 2016. Retrieval of forest above-ground biomass using multi-source data in Genhe, Inner Mongolia. Journal of Beijing Forestry University, 38(3): 64-72. [in Chinese])
李德仁, 王长委, 胡月明, 等. 2012. 遥感技术估算森林生物量的研究进展. 武汉大学学报: 信息科学版, 37(6): 631-635.
(Li D R, Wang C W, Hu Y M, et al. 2012. General review on remote sensing-based biomass estimation. Geomatics and Information Science of Wuhan University, 37(6): 631-635. [in Chinese])
李明阳, 余超, 张密芳, 等. 2015. 紫金山风景林生物量及其驱动因素时间轨迹分析. 北京林业大学学报, 37(2): 1-7.
(Li M Y, Yu C, Zhang M F, et al. 2015.Time trajectory analysis of scenic forest biomass and driving factors in Zijin Mountain, eastern China. Journal of Beijing Forestry University, 37(2): 1-7. [in Chinese])
戚玉娇, 李凤日. 2015. 基于KNN方法的大兴安岭地区森林地上碳储量遥感估算. 林业科学, 51(5): 46-55.
(Qi Y J, Li F R. 2015. Remote sensing estimation of aboveground forest carbon storage in Daxing'an Mountains based on KNN method. Scientia Silvae Sinicae, 51(5): 46-55. [in Chinese])
徐婷, 曹林, 申鑫, 等. 2015. 基于机载激光雷达与 Landsat 8 OLI 数据的亚热带森林生物量估算. 植物生态学报, 39(4): 309-321.
(Xu T, Cao L, Shen X, et al. 2015. Estimates of subtropical forest biomass based on airborne LiDAR and Landsat 8 OLI data. Chinese Journal of Plant Ecology, 39(4): 309-321. [in Chinese])
徐涵秋, 唐菲. 2013. 新一代 Landsat 系列卫星: Landsat 8 遥感影像新增特征及其生态环境意义. 生态学报, 33(11): 3249-3257.
(Xu H Q, Tang F. 2013.Analysis of new characteristics of the first Landsat 8 image and their eco-environmental significance. Acta Ecologica Sinica, 33(11): 3249-3257. [in Chinese])
Anderson G P, Felde G W, Hoke M L, et al. 2002. MODTRAN4-based atmospheric correction algorithm: FLAASH (fast line-of-sight atmospheric analysis of spectral hypercubes). AeroSense 2002, International Society for Optics and Photonics, 4725:65-71.
Bastin J F, Barbier N, Couteron P, et al. 2014. Aboveground biomass mapping of African forest mosaics using canopy texture analysis: toward a regional approach. Ecological Applications, 24(8): 1984-2001.
Birth G S, McVey G R. 1968. Measuring the color of growing turf with a reflectance spectrophotometer. Agronomy Journal, 60(6): 640-643.
Breiman L. 1996. Heuristics of instability and stabilization in model selection. The Annals of Statistics, 24(6): 2350-2383.
Breiman L. 2001. Random forests. Machine Learning, 45(1): 5-32.
Chinembiri T S, Bronsveld M C, Rossiter D G, et al. 2013. The precision of C stock estimation in the ludhikola watershed using model-based and design-based approaches. Natural Resources Research, 22(4): 297-309.
Crookston N L, Finley A O. 2008. yaImpute: an R package for kNN imputation. Journal of Statistical Software, 23 (10): 16.
Cutler M E J, Boyd D S, Foody G M, et al. 2012. Estimating tropical forest biomass with a combination of SAR image texture and Landsat TM data: An assessment of predictions between regions. ISPRS Journal of Photogrammetry and Remote Sensing, 70(3): 66-77.
Dube T, Mutanga O. 2015. Investigating the robustness of the new Landsat-8 operational land imager derived texture metrics in estimating plantation forest aboveground biomass in resource constrained areas. ISPRS Journal of Photogrammetry and Remote Sensing, 108: 12-32.
Efron B, Tibshirani R. 1986. Bootstrap methods for standard errors, confidence intervals, and other measures of statistical accuracy. Statistical Science, 1(1): 54-75.
Eckert S. 2012. Improved forest biomass and carbon estimations using texture measures from WorldView-2 satellite data. Remote Sensing, 4(4): 810-829.
Foody G M, Cutler M E, Mcmorrow J, et al. 2001. Mapping the biomass of Bornean tropical rain forest from remotely sensed data. Global Ecology and Biogeography, 10(4): 379-387.
Franco-Lopez H, Ek A R, Bauer M E. 2001. Estimation and mapping of forest stand density, volume, and cover type using the k-nearest neighbors method. Remote Sensing of Environment, 77(3): 251-274.
Gara T W, Murwira A, Chivhenge E, et al. 2014. Estimating wood volume from canopy area in deciduous woodlands of Zimbabwe. Southern Forests: A Journal of Forest Science, 76(4): 237-244.
Güneralp I, Filippi A M, Randall J. 2014. Estimation of floodplain aboveground biomass using multispectral remote sensing and nonparametric modeling. International Journal of Applied Earth Observation and Geoinformation, 33: 119-126.
Haralick R M, Shanmugam K. 1973. Textural features for image classification. IEEE Transactions on Systems, Man, and Cybernetics, 3(6): 610-621.
Huete A, Didan K, Miura T, et al. 2002. Overview of the radiometric and biophysical performance of the MODIS vegetation indices. Remote Sensing of Environment, 83(1): 195-213.
Kaufman Y J, Tanre D. 1992. Atmospherically resistant vegetation index (ARVI) for EOS-MODIS. IEEE Transactions on Geoscience and Remote Sensing, 30(2): 261-270.
Kelsey K C, Neff J C. 2014. Estimates of aboveground biomass from texture analysis of Landsat imagery. Remote Sensing, 6(7): 6407-6422.
Lee J H, Philpot W D. 1991. Spectral texture pattern matching: a classifier for digital imagery. IEEE Transactions on Geoscience and Remote Sensing, 29(4): 545-554.
Li S, Harner E J, Adjeroh D A. 2011. Random KNN feature selection-a fast and stable alternative to random forests. BMC Bioinformatics, 12(1): 1-11.
Lu D. 2005. Aboveground biomass estimation using Landsat TM data in the Brazilian Amazon. International Journal of Remote Sensing, 26(12): 2509-2525.
Lu D. 2006. The potential and challenge of remote sensing-based biomass estimation. International Journal of Remote Sensing, 27(7): 1297-1328.
Ranson K J, Sun G. 1994. Mapping biomass of a northern forest using multifrequency SAR data. IEEE Transactions on Geoscience and Remote Sensing, 32(2): 388-396.
Reese H, Nilsson M, Sandström P, et al. 2002. Applications using estimates of forest parameters derived from satellite and forest inventory data. Computers and Electronics in Agriculture, 37(1): 37-55.
Rouse J W J, Haas R H, Schell J A, et al. 1974. Monitoring vegetation systems in the great plains with ERTS. NASA Special Publication, 351: 309.
Sarker L R, Nichol J E. 2011. Improved forest biomass estimates using ALOS AVNIR-2 texture indices. Remote Sensing of Environment, 115(4): 968-977.
Sarker M L R, Nichol J, Iz H B, et al. 2013. Forest biomass estimation using texture measurements of high-resolution dual-polarization C-band SAR data. IEEE Transactions on Geoscience and Remote Sensing, 51(6): 3371-3384.
Strobl C, Boulesteix A L, Zeileis A, et al. 2007. Bias in random forest variable importance measures: illustrations, sources and a solution. BMC Bioinformatics, 8(1): 25.
Tian X, Su Z, Chen E, et al. 2012. Estimation of forest above-ground biomass using multi-parameter remote sensing data over a cold and arid area. International Journal of Applied Earth Observation and Geoinformation, 14(1): 160-168.
Tian X, Li Z, Su Z, et al. 2014. Estimating montane forest above-ground biomass in the upper reaches of the Heihe River Basin using Landsat-TM data. International Journal of Remote Sensing, 35(21): 7339-7362.
Tokola T, Pitkänen J, Partinen S, et al. 1996. Point accuracy of a non-parametric method in estimation of forest characteristics with different satellite materials. International Journal of Remote Sensing, 17(12): 2333-2351.
Troyanskaya O, Cantor M, Sherlock G, et al. 2001. Missing value estimation methods for DNA microarrays. Bioinformatics, 17(6): 520-525.
Wilson B T, Lister A J, Riemann R I. 2012. A nearest-neighbor imputation approach to mapping tree species over large areas using forest inventory plots and moderate resolution raster data. Forest Ecology and Management, 271(3): 182-198.
Wood J. 1996. The geomorphological characterisation of digital elevation models. United Kingdom:PhD thesis of University of Leicester.

[1]	曾伟生, 贺东北, 蒲莹, 肖前辉. 含地域和起源因子的马尾松立木生物量与材积方程系统[J]. 林业科学, 2019, 55(2): 75-86.
[2]	庞勇, 蒙诗栎, 李增元. 机载高分辨率遥感影像的傅氏纹理因子估测温带森林地上生物量[J]. 林业科学, 2017, 53(3): 94-104.
[3]	李兰, 陈尔学, 李增元, 任冲, 赵磊, 谷鑫志. 森林地上生物量的多基线InSAR层析估测方法[J]. 林业科学, 2017, 53(11): 85-93.
[4]	何季, 吴波, 鲍芳, 李嘉竹, 姚斌, 叶静芸, 刘建康, 辛智鸣. 人工模拟增雨对乌兰布和沙漠白刺生物量分配的影响[J]. 林业科学, 2016, 52(5): 81-91.
[5]	冯琦, 陈尔学, 李增元, 李兰, 赵磊. 基于机载P-波段全极化SAR数据的复杂地形森林地上生物量估测方法[J]. 林业科学, 2016, 52(3): 10-22.
[6]	白玉锋, 陈超群, 徐海量, 张广朋, 张沛, 凌红波. 塔里木河下游荒漠植被地上生物量空间分布与地下水埋深关系[J]. 林业科学, 2016, 52(11): 1-10.
[7]	韩新生, 邓莉兰, 王彦辉, 熊伟, 李振华, 刘千, 王艳兵, 孙浩. 六盘山叠叠沟华北落叶松人工林地上生物量的坡面变化[J]. 林业科学, 2015, 51(3): 132-139.
[8]	郭云, 李增元, 陈尔学, 田昕, 凌飞龙. 甘肃黑河流域上游森林地上生物量的多光谱遥感估测[J]. 林业科学, 2015, 51(1): 140-149.
[9]	李文梅, 陈尔学, 李增元, 赵磊. 森林地上生物量的极化干涉SAR相干层析估测方法[J]. 林业科学, 2014, 50(2): 70-77.
[10]	贺鹏;张会儒;雷相东;徐广;高祥. 基于地统计学的森林地上生物量估计[J]. , 2013, 49(5): 101-109.
[11]	徐小军;周国模;杜华强;董德进;崔瑞蕊;周宇峰;沈振明. 基于Landsat TM数据估算雷竹林地上生物量[J]. 林业科学, 2011, 47(9): 1-6.
[12]	曾伟生;夏忠胜;朱松;罗洪章. 贵州省人工马尾松立木材积和地上生物量方程研建[J]. 林业科学, 2011, 47(3): 96-101.
[13]	杨春花周小平王小明. 卧龙自然保护区华西箭竹地上生物量回归模型[J]. , 2008, 44(3): 113-123.
[14]	刘兴良;刘世荣宿以明蔡小虎马钦彦. 巴郎山川滇高山栎灌丛地上生物量及其对海拔梯度的响应[J]. 林业科学, 2006, 42(02): 1-7.