机载AISA Eagle Ⅱ高光谱数据在温带天然林树种分类中的应用
|
摘要
以内蒙古根河地区的温带天然林试验区为研究对象,采用支持向量机(SVM)方法,对经过大气校正后的机载AISA Eagle Ⅱ高光谱地表反射率影像分航带进行树种分类,将地面光谱测量与高光谱同平台的高分辨率航空相片结合,进行训练样本的选择,使用地面样地数据对分类结果进行验证。结果表明:利用AISA Eagle Ⅱ高光谱影像对温带天然林区分类的总体精度和kappa系数分别达到了96.71%和0.95;灌木分类精度最高,其制图精度和用户精度分别达到了98.07%和98.31%;落叶松和白桦的用户精度分别为98%和94%。
With the temperate forest area of Genhe in Inner Mongolia as an example, Support Vector Machine(SVM) was used to classify the reflectance images. Tree species training samples were selected based on high resolution aerial photographs. The filed data were used to verify the classification results. The overall accuracy and Kappa coefficient reached 96.71% and 0.95, respectively, and for a single tree species it had good classification result. The shrub had highest classification accuracy, with the mapping accuracy and user accuracy of 98.07% and 98.31%, respectively. The user accuracies of Larix gmelinii and Betula platyphylla were 98% and 94%, respectively.
With the temperate forest area of Genhe in Inner Mongolia as an example, Support Vector Machine(SVM) was used to classify the reflectance images. Tree species training samples were selected based on high resolution aerial photographs. The filed data were used to verify the classification results. The overall accuracy and Kappa coefficient reached 96.71% and 0.95, respectively, and for a single tree species it had good classification result. The shrub had highest classification accuracy, with the mapping accuracy and user accuracy of 98.07% and 98.31%, respectively. The user accuracies of Larix gmelinii and Betula platyphylla were 98% and 94%, respectively.
引文
[1] 陶江玥,刘丽娟,庞勇.基于机载激光雷达和高光谱数据的树种识别方法[J].浙江农林大学学报,2018,35(2):314-323.
[2] 于丽柯,于颖,柳向宇.基于高光谱影像的树种分类[J].东北林业大学学报,2016,44(9):40-44.
[3] VAN AARDT J A,NORRIS-ROGERS M.Spectral-age interactions in managed,even-aged eucalyptus plantations:application of discriminant analysis and classification and regression trees approaches to hyperspectral data[J].International Journal of Remote Sensing,2008,29(6):1841-1845.
[4] DALPONTE M,BRUZZONE L,GIANELLE D.Tree species classification in the Southern Alps based on the fusion of very high geometrical resolution multispectral/hyperspectral images and LiDAR data[J].Remote Sensing of Environment,2012,123:258-270.
[5] DIAN Y Y,LI Z Y,PANG Y.Spectral and texture features combined for forest tree species classification with airborne hyperspectral imagery[J].Journal of the Indian Society of remote Sensing,2015,43(1):101-107.
[6] 刘怡君,庞勇,廖声熙,等.机载LiDAR和高光谱融合实现普洱山区树种分类[J].林业科学研究,2016,29(3):407-412.
[7] FéRET J,ASNER G P.Tree Species discrimination in tropical forests using airborne imaging spectroscopy[J].IEEE Transactions on Geoscience and Remote Sensing,2013,51(1):73-84.
[8] 刘丽娟,庞勇,范文义,等.基于机载激光雷达和高光谱融合的温带复杂森林树种识别[J].遥感学报,2013,17(3):679-695.
[9] 荚文,庞勇,岳彩荣,等.基于AISA Eagle II机载高光谱数据的普洱市山区森林分类[J].林业调查规划,2015,40(1):9-14.
[10] KABIR Y P,ONISIMO M,RIYAD I.Commercial tree species d.iscrimination using airborne AISA Eagle hyperspectral imagery and partial least squares discriminant analysis (PLS-DA)in Kwa Zulu-Natal,South Africa[J].ISPRS Journal of Photogrammetry and Remote Sensing,2015,79:19-28.doi.org/10.1016/j.isprsjprs.2013.01.013.
[11] NAIDOO L,CHO M A,MATHIEU R,et al.Classification of savanna tree species in the Greater Kruger National Park region by integrating hyperspectral and LiDAR data in a Random Forest data mining environment[J].Isprs Journal of Photogrammetry and Remote Sensing,2012,69:167-179.doi:10.1016/j.isprsjprs.2012.03.005.
[12] VLADIMIR V K,DMITRIEV E V.Testing different classification methods in airborne hyperspectral imagery processing[J].Optics Express,2016,24(10):956-965.doi:10.1364/OE.24.00A956.
[13] VLADIMIR V K,DMITRIEV E V,SOKOLOV A A.Improved technique for retrieval of forest parameters from hyperspectral remote sensing data[J].Optics Express,2015,23(24):1342-1353.doi:10.1364/OE.23.0A1342.
[14] MERCIER G,LENNON M.Support Vector Machines for Hyperspectral Image Classification with Spectral-Based Kernels[J].IEEE International Geoscience and Remote Sensing Symposium,IGARSS,2003,3:21-25.doi:10.1109/IGARSS.2003.1293752.
[15] PANG Y,LI Z,JU H.LiCHy:The CAF’s LiDAR,CCD and Hyperspectral Integrated Airborne Observation System[J].Remote Sensing,2016,8(5).doi:10.3390/rs8050398.
[16] BAZI Y,MELGANI F.Toward an optimal SVM classification system for hyperspectral remote sensing images[J].IEEE Trans Geosci Remote Sens,2006,44(11):3374-3385.doi:10.1109/tgrs.2006.880628.
[17] 胡国胜,钱玲,张国红.支持向量机的多分类算法[J].系统工程与电子技术,2006,28(1):127-132.
[18] 林娜,杨武年,王斌.基于核最小噪声分离变换的高光谱遥感影像多类SVM分类[J].计算机应用与软件,2014,31(6):116-119.
[2] 于丽柯,于颖,柳向宇.基于高光谱影像的树种分类[J].东北林业大学学报,2016,44(9):40-44.
[3] VAN AARDT J A,NORRIS-ROGERS M.Spectral-age interactions in managed,even-aged eucalyptus plantations:application of discriminant analysis and classification and regression trees approaches to hyperspectral data[J].International Journal of Remote Sensing,2008,29(6):1841-1845.
[4] DALPONTE M,BRUZZONE L,GIANELLE D.Tree species classification in the Southern Alps based on the fusion of very high geometrical resolution multispectral/hyperspectral images and LiDAR data[J].Remote Sensing of Environment,2012,123:258-270.
[5] DIAN Y Y,LI Z Y,PANG Y.Spectral and texture features combined for forest tree species classification with airborne hyperspectral imagery[J].Journal of the Indian Society of remote Sensing,2015,43(1):101-107.
[6] 刘怡君,庞勇,廖声熙,等.机载LiDAR和高光谱融合实现普洱山区树种分类[J].林业科学研究,2016,29(3):407-412.
[7] FéRET J,ASNER G P.Tree Species discrimination in tropical forests using airborne imaging spectroscopy[J].IEEE Transactions on Geoscience and Remote Sensing,2013,51(1):73-84.
[8] 刘丽娟,庞勇,范文义,等.基于机载激光雷达和高光谱融合的温带复杂森林树种识别[J].遥感学报,2013,17(3):679-695.
[9] 荚文,庞勇,岳彩荣,等.基于AISA Eagle II机载高光谱数据的普洱市山区森林分类[J].林业调查规划,2015,40(1):9-14.
[10] KABIR Y P,ONISIMO M,RIYAD I.Commercial tree species d.iscrimination using airborne AISA Eagle hyperspectral imagery and partial least squares discriminant analysis (PLS-DA)in Kwa Zulu-Natal,South Africa[J].ISPRS Journal of Photogrammetry and Remote Sensing,2015,79:19-28.doi.org/10.1016/j.isprsjprs.2013.01.013.
[11] NAIDOO L,CHO M A,MATHIEU R,et al.Classification of savanna tree species in the Greater Kruger National Park region by integrating hyperspectral and LiDAR data in a Random Forest data mining environment[J].Isprs Journal of Photogrammetry and Remote Sensing,2012,69:167-179.doi:10.1016/j.isprsjprs.2012.03.005.
[12] VLADIMIR V K,DMITRIEV E V.Testing different classification methods in airborne hyperspectral imagery processing[J].Optics Express,2016,24(10):956-965.doi:10.1364/OE.24.00A956.
[13] VLADIMIR V K,DMITRIEV E V,SOKOLOV A A.Improved technique for retrieval of forest parameters from hyperspectral remote sensing data[J].Optics Express,2015,23(24):1342-1353.doi:10.1364/OE.23.0A1342.
[14] MERCIER G,LENNON M.Support Vector Machines for Hyperspectral Image Classification with Spectral-Based Kernels[J].IEEE International Geoscience and Remote Sensing Symposium,IGARSS,2003,3:21-25.doi:10.1109/IGARSS.2003.1293752.
[15] PANG Y,LI Z,JU H.LiCHy:The CAF’s LiDAR,CCD and Hyperspectral Integrated Airborne Observation System[J].Remote Sensing,2016,8(5).doi:10.3390/rs8050398.
[16] BAZI Y,MELGANI F.Toward an optimal SVM classification system for hyperspectral remote sensing images[J].IEEE Trans Geosci Remote Sens,2006,44(11):3374-3385.doi:10.1109/tgrs.2006.880628.
[17] 胡国胜,钱玲,张国红.支持向量机的多分类算法[J].系统工程与电子技术,2006,28(1):127-132.
[18] 林娜,杨武年,王斌.基于核最小噪声分离变换的高光谱遥感影像多类SVM分类[J].计算机应用与软件,2014,31(6):116-119.