SVM-based forest mapping of Wolong Giant Panda Habitat using SAR data
ZHOU Xiaoyu1, 2, CHEN Fulong1, 3, JIANG Aihui4
1. Key Laboratory of Digital Earth Science, Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences, Beijing 100094, China; 2. University of Chinese Academy of Sciences, Beijing 100049, China; 3. International Centre on Space Technologies for Natural and Cultural Heritage Under the Auspices of UNESCO, Beijing 100094, China; 4. College of Geomatics, Shandong University of Science and Technology, Qingdao 266590, China
Abstract：Generally, the conservation of Wolong Giant Panda Habitat (Natural World Heritage site) is significant for the sustainability of the rare species of Giant Panda. As we know, forest coverage can be an essential impact on the suitability of the habitat. Owing to the all-weather, all-day operation capability of radar systems, in this study, the authors investigated the performance of Synthetic Aperture Radar (SAR) images in fine mapping of forests using multi-temporal/polarization PALSAR data. The authors firstly corrected the radiometric distortion of SAR data induced by the cliffy topography; then the authors selected 5 different temporal acquisitions for the forest mapping using the Support Vector Machine (SVM) approach. 5 multi-temporal/dual-polarization indexes, i.e., HHm,HVm,TSD,HHm-HVm and HHm/HVm, were applied for the training and classification. Experimental results demonstrated that the combination of HHm, HVm, TSD and HHm-HVm derived an optimal classification (e.g., total accuracy and user accuracy of forest/non-forest were 86.90%,82.34% and 92.83%, respectively), better than the single-temporal/polarization mode (total classification accuracy of 55.47%). This study shows the effectiveness of multi-temporal/polarization SAR data in forest fine mapping, particularly in the monitoring and evaluation of natural heritage sites located in cloudy and rainy environments.
[1] 孙克勤.中国的世界自然遗产战略管理研究[J].中国人口·资源与环境,2011,21(3):547-550. Sun K Q.Study on strategic management of world natural heritage in China[J].China Population,Resources and Environment,2011,21(3):547-550. [2] 申国珍,谢宗强,冯朝阳,等.汶川地震对大熊猫栖息地的影响与恢复对策[J].植物生态学报,2008,32(6):1417-1425. Shen G Z,Xie Z Q,Feng C Y,et al.Influence of the Wenchuan earthquake on giant panda habitats and strategies for restoration[J].Journal of Plant Ecology,2008,32(6):1417-1425. [3] Malhi Y,Roberts J T,Betts R A,et al.Climate change,deforestation,and the fate of the Amazon[J].Science,2008,319(5860):169-172. [4] Avtar R,Takeuchi W,Sawada H.Full polarimetric PALSAR-based land cover monitoring in Cambodia for implementation of REDD policies[J].International Journal of Digital Earth,2013,6(3):255-275. [5] 严婷婷,边红枫,廖桂项,等.森林湿地遥感信息提取方法研究现状[J].国土资源遥感,2014,26(2):11-18.doi:10.6046/gtzyyg.2014.02.03"> doi:10.6046/gtzyyg.2014.02.03. Yan T T,Bian H F,Liao G X,et al.Research status of methods for mapping forested wetlands based on remote sensing[J].Remote Sensing for Land and Resources,2014,26(2):11-18.doi:10.6046/gtzyyg.2014.02.03"> doi:10.6046/gtzyyg.2014.02.03. [6] Schlund M,von Poncet F,Hoekman D H,et al.Importance of bistatic SAR features from TanDEM-X for forest mapping and monitoring[J].Remote Sensing of Environment,2014,151:16-26. [7] Rahman M M,Tetuko Sri Sumantyo J.Quantifying deforestation in the Brazilian Amazon using Advanced Land Observing Satellite Phased Array L-band Synthetic Aperture Radar(ALOS PALSAR) and Shuttle Imaging Radar(SIR)-C data[J].Geocarto International,2012,27(6):463-478. [8] Næsset E,Gobakken T,Solberg S,et al.Model-assisted regional forest biomass estimation using LiDAR and InSAR as auxiliary data:A case study from a boreal forest area[J].Remote Sensing of Environment,2011,115(12):3599-3614. [9] Almeida-Filho R,Rosenqvist A,Shimabukuro Y E,et al.Detecting deforestation with multitemporal L-band SAR imagery:A case study in western Brazilian Amazônia[J].International Journal of Remote Sensing,2007,28(6):1383-1390. [10] Santoro M,Fransson J E S,Eriksson L E B,et al.Clear-cut detection in Swedish boreal forest using multi-temporal ALOS PALSAR backscatter data[J].IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2010,3(4):618-631. [11] Whittle M,Quegan S,Uryu Y,et al.Detection of tropical deforestation using ALOS-PALSAR:A Sumatran case study[J].Remote Sensing of Environment,2012,124:83-98. [12] 李 亮,应国伟,文学虎,等.融合时间特征的高分辨率遥感影像分类[J].国土资源遥感,2016,28(2):91-98.doi:10.6046/gtzyyg.2016.02.15"> doi:10.6046/gtzyyg.2016.02.15. Li L,Ying G W,Wen X H,et al.Classification of high spatial resolution remotely sensed images by temporal feature fusion[J].Remote Sensing for Land and Resources,2016,28(2):91-98.doi:10.6046/gtzyyg.2016.02.15"> doi:10.6046/gtzyyg.2016.02.15. [13] Song M J,Civico D.Road extraction using SVM and image segmentation[J].Photogrammetric Engineering and Remote Sensing,2004,70(12):1365-1371. [14] Brown M,Lewis H G,Gunn S R.Linear spectral mixture models and support vector machines for remote sensing[J].IEEE Transactions on Geoscience and Remote Sensing,2000,38(5):2346-2360. [15] Attarchi S,Gloaguen R.Improving the estimation of above ground biomass using dual polarimetric PALSAR and ETM + data in the Hyrcanian mountain forest(Iran)[J].Remote Sensing,2014,6(5):3693-3715. [16] Attarchi S,Gloaguen R.Classifying complex mountainous forests with L-band SAR and Landsat data integration:A comparison among different machine learning methods in the Hyrcanian forest[J].Remote Sensing,2014,6(5):3624-3647. [17] 国家林业局.全国第三次大熊猫调查报告[M].北京:科学出版社,2006:26. The State Forestry Administration of the People’s Republic of China.The 3rd National Survey Report on Giant Panda in China[M].Beijing:Science Publishing House,2006:26. [18] 周世强,黄金燕,张亚辉,等.卧龙自然保护区大熊猫栖息地植物群落多样性Ⅴ:不同竹林的物种多样性[J].应用与环境生物学报,2009,15(2):180-187. Zhou S Q,Huang J Y,Zhang Y H,et al.Diversity of plant community of giant Panda’s habitat in the Wolong Nature Reserve Ⅴ:Species diversity in different bamboo forests[J].Chinese Journal of Applied & Environmental Biology,2009,15(2):180-187. [19] Castel T,Beaudoin A,Stach N,et al.Sensitivity of space-borne SAR data to forest parameters over sloping terrain.Theory and experiment[J].International Journal of Remote Sensing,2001,22(12):2351-2376. [20] Ulander L M.Radiometric slope correction of synthetic-aperture Radar images[J].IEEE Transactions on Geoscience and Remote Sensing,1996,34(5):1115-1122. [21] Lucas R,Armston J,Fairfax R,et al.An evaluation of the ALOS PALSAR L-band backscatter-above ground biomass relationship Queensland,Australia:Impacts of surface moisture condition and vegetation structure[J].IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2010,3(4):576-593. [22] Santoro M,Fransson J E S,Eriksson L E B,et al.Signatures of ALOS PALSAR L-band backscatter in Swedish forest[J].IEEE Transactions on Geoscience and Remote Sensing,2009,47(12):4001-4019. [23] Powell R L,Matzke N,de Souza C Jr,et al.Sources of error in accuracy assessment of thematic land-cover maps in the Brazilian Amazon[J].Remote Sensing of Environment,2004,90(2):221-234. [24] Dong J W,Xiao X M,Sheldon S,et al.A 50-m forest cover map in Southeast Asia from ALOS/PALSAR and its application on forest fragmentation assessment[J].PLoS One,2014,9(1):e85801. [25] Dong J W,Xiao X M,Sheldon S,et al.A comparison of forest cover maps in Mainland Southeast Asia from multiple sources:PALSAR,MERIS,MODIS and FRA[J].Remote Sensing of Environment,2012,127:60-73. [26] Cortes C,Vapnik V.Support-vector networks[J].Machine Learning,1995,20(3):273-297. [27] Vapnik V N.The Nature of Statistical Learning Theory[M].2nd ed.New York:Springer,2000. [28] 陈 波,张友静,陈 亮.结合纹理的SVM遥感影像分类研究[J].测绘工程,2007,16(5):23-27. Chen B,Zhang Y J,Chen L.RS Image classification based on SVM method with texture[J].Engineering of Surveying and Mapping,2007,16(5):23-27. [29] 奉国和.SVM分类核函数及参数选择比较[J].计算机工程与应用,2011,47(3):123-124,128. Feng G H.Parameter optimizing for Support Vector Machines classification[J].Computer Engineering and Applications,2011,47(3):123-124,128. [30] Chang C C,Lin C J.LIBSVM:A library for support vector machines[EB/OL].http://www.csie.ntu.edu.tw/~cjlin/libsvm/. [31] Hsu C W,Chang C C,Lin C J.A practical guide to support vector classification[EB/OL].[2009-06-20].http://www.csie.ntu.edu.tw/~cjlin/papers/guide/guide.pdf. [32] Miettinen J,Liew S C.Separability of insular Southeast Asian woody plantation species in the 50 m resolution ALOS PALSAR mosaic product[J].Remote Sensing Letters,2011,2(4):299-307.
2017, Vol. 29 
