基于随机森林算法的赣南柑橘果园遥感信息提取
|
摘要
选择春、秋季低云量Landsat-8卫星遥感影像,构建包含有多光谱地表反射率、光谱指数、几何纹理和地形因子的分类特征集,通过随机森林分类算法开展赣南柑橘果园空间分布遥感制图研究.研究结果表明:利用春季影像提取的柑橘果园整体精度为91.12%,Kappa系数为0.88,优于秋季影像提取结果;随机森林算法在赣南柑橘果园识别制图中具有较高的分类精度和较好的适用性,利用降维的分类特征提取柑橘果园也具有较高精度;赣南柑橘果园面积约1 794.26 km~2,具有一定比例的陡坡种植现象,寻乌、信丰、安远等3县的柑橘果园呈现规模化、连片化的景观.
Several Landsat OLI images acquired in spring and autumn are selected to map citrus orchards distribution. Random Forest( RF) classifier is utilized to implement a supervised classification on the dataset including multi-spectral reflectance,vegetation and moisture indices,texture information and topographic features. The results show that classification with spring image is successful with an overall accuracy( OA) of 91. 12% and a Kappa statistic of 0. 88. And it is superior to that with autumn images. RF is highly suitable for the identi-fication and classification of citrus orchards. And classification with an optimal subset of discrimination features is also acceptable with a high accuracy. The area of citrus orchards is about 1 794. 26 km~2 and a certain proportion of citrus orchards is cultivated on steep slopes. The landscape characteristics of citrus orchards in some counties such as Xunwu,Xinfeng and Anyuan became single,continuous and massive.
Several Landsat OLI images acquired in spring and autumn are selected to map citrus orchards distribution. Random Forest( RF) classifier is utilized to implement a supervised classification on the dataset including multi-spectral reflectance,vegetation and moisture indices,texture information and topographic features. The results show that classification with spring image is successful with an overall accuracy( OA) of 91. 12% and a Kappa statistic of 0. 88. And it is superior to that with autumn images. RF is highly suitable for the identi-fication and classification of citrus orchards. And classification with an optimal subset of discrimination features is also acceptable with a high accuracy. The area of citrus orchards is about 1 794. 26 km~2 and a certain proportion of citrus orchards is cultivated on steep slopes. The landscape characteristics of citrus orchards in some counties such as Xunwu,Xinfeng and Anyuan became single,continuous and massive.
引文
[1]徐艳霞.地方特色农业产业化发展问题研究[J].全国商情,2016(22):50-51.
[2]李文华,刘某承,闵庆文.中国生态农业的发展与展望[J].资源科学,2010,32(6):1015-1021.
[3]Yuan Huili,Luo Juhua,Ma Ronghua.Mapping orchards on plain terrains using multi-temporal medium-resolution satellite imagery[J].Applied Engineering in Agriculture,2015,31(3):351-362.
[4]Reis S,Ta爧demir K.Identification of hazelnut fields using spectral and Gabor textural features[J].Isprs Journal of Photogrammetry and Remote Sensing,2011,66(5):652-661.
[5]Wang Hong,Zhao Yu,Pu Ruiliang,et al.Mapping Robinia Pseudoacacia forest health conditions by using combined spectral,spatial,and textural information extracted from IKONOS imagery and random forest classifier[J].Remote Sensing,2015,7(7):9020-9044.
[6]Ma Lei,Cheng Liang,Han Wenquan,et al.Cultivated land information extraction from high-resolution unmanned aerial vehicle imagery data[J].Journal of Applied Remote Sensing,2014,8:1-25.
[7]梁守真,陈劲松,吴炳方,等.应用面向对象的决策树模型提取橡胶林信息[J].遥感学报,2015,19(3):485-494.
[8]Otukei J R,Blaschke T,Woldai T,et al.Land cover change assessment using decision trees,support vector machines and maximum likelihood classification algorithms[J].International Journal of Applied Earth Observation and Geoinformation,2010,12(1):27-31.
[9]Chakraborty A,Sachdeva K,Joshi P K.Mapping long-term land use and land cover change in the central Himalayan region using a tree-based ensemble classification approach[J].Applied Geography,2016,74:136-150.
[10]张安定,彭笃明,李德一,等.基于TM影像的果园空间信息提取技术研究[J].测绘科学,2007,32(5):121-123.
[11]于新洋,张安定,侯西勇.胶东半岛果园TM影像信息的提取决策树方法[J].测绘科学,2012,37(4):57-60.
[12]罗卫,况润元.利用环境卫星影像的东江源地区果园信息提取[J].测绘科学,2014,39(8):135-139.
[13]Diao Chunyuan,Wang Le.Incorporating plant phenological trajectory in exotic saltcedar detection with monthly time series of Landsat imagery[J].Remote Sensing of Environment,2016,182:60-71.
[14]Belgiu M,Drǎgu爫L.Random forest in remote sensing:a review of applications and future directions[J].Isprs Journal of Photogrammetry and Remote Sensing,2016,114:24-31.
[15]Tatsumi K,Yamashiki Y,Torres M A C,et al.Crop classification of upland fields using random forest of time-series Landsat 7 ETM+data[J].Computers and Electronics in Agriculture,2015,115:171-179.
[16]顾海燕,闫利,李海涛,等.基于随机森林的地理要素面向对象自动解译方法[J].武汉大学学报:信息科学版,2016,41(2):228-234.
[17]孙永明,叶川,王学雄,等.赣南脐橙果园水土流失现状调查分析[J].水土保持研究,2014,21(2):67-71.
[18]Rouse J W.Monitoring the vernal advancement and retrogradation(greenwave effect)of natural vegetation[M].College Station:Texas A M University,1974.
[19]Liu Huiqing,Huete A.A feedback based modification of the NDVI to minimize canopy background and atmospheric noise[J].IEEE Transactions on Geoscience and Remote Sensing,1995,33(2):457-465.
[20]Wilson E H,Sader S A.Detection of forest harvest type using multiple dates of Landsat TM imagery[J].Remote Sensing of Environment,2002,80(3):385-396.
[21]Gao B C.NDWI:a normalized difference water index for remote sensing of vegetation liquid water from space[J].Remote Sensing of Environment,1996,58(3):257-266.
[22]Henebry G M,Rieck D R.Applying principal components analysis to image time series:effects on scene segmentation and spatial structure[J].Remote Sensing for a Sustainable Future,1996,1(1):448-450.
[23]Haralick R M,Shanmugam K,Dinstein I H.Texture features for image classification[J].IEEE Transactions on Systems Man and Cybernetics,1973,3(6):610-621.
[24]彭燕,何国金,张兆明,等.赣南稀土矿开发区生态环境遥感动态监测与评估[J].生态学报,2016,36(6):1676-1685.
[25]Breiman L.Random forests[J].Machine Learning,2001,45(1):5-32.
[26]马玥,姜琦刚,孟治国,等.基于随机森林算法的农耕区土地利用分类研究[J].农业机械学报,2016,47(1):297-303.
[27]Waske B,Sebastian V D L,Oldenburg C,et al.Image RFA user-oriented implementation for remote sensing image analysis with Random Forests[J].Environmental Modelling and Software,2012,35:192-193.
[28]李自茂,钟八莲,孙剑斌.赣南脐橙产业发展报告(2013)[M].北京:经济管理出版社,2014.
[2]李文华,刘某承,闵庆文.中国生态农业的发展与展望[J].资源科学,2010,32(6):1015-1021.
[3]Yuan Huili,Luo Juhua,Ma Ronghua.Mapping orchards on plain terrains using multi-temporal medium-resolution satellite imagery[J].Applied Engineering in Agriculture,2015,31(3):351-362.
[4]Reis S,Ta爧demir K.Identification of hazelnut fields using spectral and Gabor textural features[J].Isprs Journal of Photogrammetry and Remote Sensing,2011,66(5):652-661.
[5]Wang Hong,Zhao Yu,Pu Ruiliang,et al.Mapping Robinia Pseudoacacia forest health conditions by using combined spectral,spatial,and textural information extracted from IKONOS imagery and random forest classifier[J].Remote Sensing,2015,7(7):9020-9044.
[6]Ma Lei,Cheng Liang,Han Wenquan,et al.Cultivated land information extraction from high-resolution unmanned aerial vehicle imagery data[J].Journal of Applied Remote Sensing,2014,8:1-25.
[7]梁守真,陈劲松,吴炳方,等.应用面向对象的决策树模型提取橡胶林信息[J].遥感学报,2015,19(3):485-494.
[8]Otukei J R,Blaschke T,Woldai T,et al.Land cover change assessment using decision trees,support vector machines and maximum likelihood classification algorithms[J].International Journal of Applied Earth Observation and Geoinformation,2010,12(1):27-31.
[9]Chakraborty A,Sachdeva K,Joshi P K.Mapping long-term land use and land cover change in the central Himalayan region using a tree-based ensemble classification approach[J].Applied Geography,2016,74:136-150.
[10]张安定,彭笃明,李德一,等.基于TM影像的果园空间信息提取技术研究[J].测绘科学,2007,32(5):121-123.
[11]于新洋,张安定,侯西勇.胶东半岛果园TM影像信息的提取决策树方法[J].测绘科学,2012,37(4):57-60.
[12]罗卫,况润元.利用环境卫星影像的东江源地区果园信息提取[J].测绘科学,2014,39(8):135-139.
[13]Diao Chunyuan,Wang Le.Incorporating plant phenological trajectory in exotic saltcedar detection with monthly time series of Landsat imagery[J].Remote Sensing of Environment,2016,182:60-71.
[14]Belgiu M,Drǎgu爫L.Random forest in remote sensing:a review of applications and future directions[J].Isprs Journal of Photogrammetry and Remote Sensing,2016,114:24-31.
[15]Tatsumi K,Yamashiki Y,Torres M A C,et al.Crop classification of upland fields using random forest of time-series Landsat 7 ETM+data[J].Computers and Electronics in Agriculture,2015,115:171-179.
[16]顾海燕,闫利,李海涛,等.基于随机森林的地理要素面向对象自动解译方法[J].武汉大学学报:信息科学版,2016,41(2):228-234.
[17]孙永明,叶川,王学雄,等.赣南脐橙果园水土流失现状调查分析[J].水土保持研究,2014,21(2):67-71.
[18]Rouse J W.Monitoring the vernal advancement and retrogradation(greenwave effect)of natural vegetation[M].College Station:Texas A M University,1974.
[19]Liu Huiqing,Huete A.A feedback based modification of the NDVI to minimize canopy background and atmospheric noise[J].IEEE Transactions on Geoscience and Remote Sensing,1995,33(2):457-465.
[20]Wilson E H,Sader S A.Detection of forest harvest type using multiple dates of Landsat TM imagery[J].Remote Sensing of Environment,2002,80(3):385-396.
[21]Gao B C.NDWI:a normalized difference water index for remote sensing of vegetation liquid water from space[J].Remote Sensing of Environment,1996,58(3):257-266.
[22]Henebry G M,Rieck D R.Applying principal components analysis to image time series:effects on scene segmentation and spatial structure[J].Remote Sensing for a Sustainable Future,1996,1(1):448-450.
[23]Haralick R M,Shanmugam K,Dinstein I H.Texture features for image classification[J].IEEE Transactions on Systems Man and Cybernetics,1973,3(6):610-621.
[24]彭燕,何国金,张兆明,等.赣南稀土矿开发区生态环境遥感动态监测与评估[J].生态学报,2016,36(6):1676-1685.
[25]Breiman L.Random forests[J].Machine Learning,2001,45(1):5-32.
[26]马玥,姜琦刚,孟治国,等.基于随机森林算法的农耕区土地利用分类研究[J].农业机械学报,2016,47(1):297-303.
[27]Waske B,Sebastian V D L,Oldenburg C,et al.Image RFA user-oriented implementation for remote sensing image analysis with Random Forests[J].Environmental Modelling and Software,2012,35:192-193.
[28]李自茂,钟八莲,孙剑斌.赣南脐橙产业发展报告(2013)[M].北京:经济管理出版社,2014.