基于Google Earth Engine与多源遥感数据的海南水稻分类研究
|
摘要
水稻是中国乃至亚洲的重要粮食作物之一,稻米产量关系到民生福祉。及时、准确的水稻分布信息是监测水稻产量、调控农业资源配给的基础。遥感(Remote Sensing)技术能够提供大范围地表的时间序列光谱变化特征,常用于大尺度范围的作物监测。然而,传统基于水稻生长关键时期光谱特征的分类、提取方法对遥感数据的时间分辨率要求较高。由于我国南方水稻产区湿热,雨季云污染降低了遥感数据的有效时间分辨率,因此上述方法在该地难以推广。融合多源遥感数据的分类方案变相缩短了卫星的重访周期,使多云气候区基于遥感影像的水稻分类成为可能。然而,集成多源数据所需更高的数据处理效率和存储需求也成为限制省级乃至更大范围水稻分类的主要因素。本研究基于谷歌地球引擎(Google Earth Engine)云平台,在线调用中分辨率的光学、微波遥感数据,创新性地采用了按月提取、按直方图大小提取特征的方式,采用随机森林分类器,绘制海南省2016年10 m分辨率水稻种植分布图。实验结果证明,该方法可以用于南方多云地区水稻分类,提取结果能够体现不同地类之间的差异,且与实际地表的地块边界、纹理符合良好。经过地表样本点的验证,总体精度为93.2%,满足实际应用需求。因此,本研究采用的自动分类流程能够准确、高效地提取海南省的水稻种植范围,可以向其他地区大范围推广。
Rice is one of the main grain crops in China and East Asia, including China. The annual yield of rice has a significant influence on domestic livelihood. Therefore, timely and accurate assessment of rice distribution information is crucial for forecasting rice yields and optimize the allocation of agricultural resources. Remote sensing(RS) images can provide time series surface spectral, and other electro-magnetic, dynamics over a largescale land surface, which are commonly used for large-scale crop monitoring. However, routine rice classifying strategies provided by the RS images during key growth stages, require spectral patterns at high frequency. This method appears to be impractical within South China, as the number of high quality RS images are difficult to obtain due to cloud contamination caused by the hot and wet weather. A combination of various RS images of rice classification from multi-platforms provide an indirect way of reducing the revisit period in routine rice classification, thus enabling successful crop mapping in cloudy regions. However, this causes difficulty with data manipulation and storage, especially when conducting classification work at province or large area levels. To address these issues, this research utilizes Google Earth Engine(a cloud-based geospatial analysis platform running on the Google server)to collect online optic RS data and micro-wave RS data at diverse resolutions for rice mapping. A distribution map of paddy rice at 10-m spatial resolution in the Hainan Province in 2016 was made by using the combined methods of random forest(RF) classification and a pattern-matching strategy based on conjunct features extracted at monthly level and histogram value distribution. Results showed this method was suitable for rice mapping in Hainan and could show clear feature divergence between the different land surface cover types. Spatial distribution results corresponded well with the actual edges of the field, along with texture information. The rice classification result of the Hainan Province was validated using sample points captured on the ground and achieved overall accuracy of 93.2%, indicating reliability for practical application.Overall, the automatic rice classifying strategy was able to map paddy rice with high efficiency and sufficient accuracy in the Hainan Province, and could be applied to other vast areas.
Rice is one of the main grain crops in China and East Asia, including China. The annual yield of rice has a significant influence on domestic livelihood. Therefore, timely and accurate assessment of rice distribution information is crucial for forecasting rice yields and optimize the allocation of agricultural resources. Remote sensing(RS) images can provide time series surface spectral, and other electro-magnetic, dynamics over a largescale land surface, which are commonly used for large-scale crop monitoring. However, routine rice classifying strategies provided by the RS images during key growth stages, require spectral patterns at high frequency. This method appears to be impractical within South China, as the number of high quality RS images are difficult to obtain due to cloud contamination caused by the hot and wet weather. A combination of various RS images of rice classification from multi-platforms provide an indirect way of reducing the revisit period in routine rice classification, thus enabling successful crop mapping in cloudy regions. However, this causes difficulty with data manipulation and storage, especially when conducting classification work at province or large area levels. To address these issues, this research utilizes Google Earth Engine(a cloud-based geospatial analysis platform running on the Google server)to collect online optic RS data and micro-wave RS data at diverse resolutions for rice mapping. A distribution map of paddy rice at 10-m spatial resolution in the Hainan Province in 2016 was made by using the combined methods of random forest(RF) classification and a pattern-matching strategy based on conjunct features extracted at monthly level and histogram value distribution. Results showed this method was suitable for rice mapping in Hainan and could show clear feature divergence between the different land surface cover types. Spatial distribution results corresponded well with the actual edges of the field, along with texture information. The rice classification result of the Hainan Province was validated using sample points captured on the ground and achieved overall accuracy of 93.2%, indicating reliability for practical application.Overall, the automatic rice classifying strategy was able to map paddy rice with high efficiency and sufficient accuracy in the Hainan Province, and could be applied to other vast areas.
引文
[1] Elert E. Rice by the numbers:A good grain[J]. Nature,2014,514(7524):50-51.
[2] Khush G S. What it will take to feed 5.0 billion rice consumers in 2030[J]. Plant Molecular Biology, 2005,59(1):1-6.
[3] Zhang Q, Zhang W, Li T, et al. Projective analysis of staple food crop productivity in adaptation to future climate change in china[J]. International Journal of Biometeorology, 2017,61(8):1-16.
[4] Bouman, B. How much water does rice use[J]? Rice Today, 2009,69:115-133.
[5] Tao F, Hayashi Y, Zhang, Z, Sakamoto T, et al. Global warming, rice production, and water use in china:Developing a probabilistic assessment[J]. Agricultural&Forest Meteorology, 2008,148(1):94-110.
[6] Knauer K, Knauer K. Remote sensing of rice crop areas[J]. International Journal for Remote Sensing, 2013,34(6):2101-2139.
[7] Ramankutty N, Foley J. Characterizing patterns of global land use:An analysis of global croplands data[J]. Global Biogeochemical Cycles, 1998,12(4):667-685.
[8] Ramankutty N, Foley J A. Estimating historical changes in global land cover:Croplands from 1700 to 1992[J].Global Biogeochemical Cycles, 1999,13(4):997-1027.
[9] Li P, Feng Z M, Jiang, et al. Changes in rice cropping systems in the poyang lake region, china during 2004-2010[J].Journal of Geographical Sciences, 2012,22(4):653-668.
[10] Thenkabail P S. Mapping rice areas of south asia using modis multitemporal data[J]. Journal of Applied Remote Sensing, 2011,5(1):863-871.
[11]黄青,吴文斌,邓辉,等.2009年江苏省冬小麦和水稻种植面积信息遥感提取及长势监测[J].江苏农业科学,2016(6):508-511.[Huang Q, Wu W B, Deng H, et al. Mapping area and condition monitoring of winter wheat and rice in JiangSu province in 2009[J]. Jiangsu Agricultural Sciences, 2016(6):508-511.]
[12] Singha M, Wu B, Zhang M. An object-based paddy rice classification using multi-spectral data and crop phenology in assam, northeast india[J]. Remote Sensing, 2016,8(6):479.
[13] Xiao X, Boles S, Liu J, et al. Mapping paddy rice agriculture in southern china using multi-temporal modis images[J]. Remote Sensing of Environment, 2005,95(4):480-492.
[14] Zhang X, Zhang M, Zheng Y, et al. Crop mapping using proba-v time series data at the yucheng and hongxing farm in china[J]. Remote Sensing, 2016,8(11):915.
[15] Choudhury I, Chakraborty M. Sar signature investigation of rice crop using radarsat data[J]. International Journal of Remote Sensing, 2006,27(3):519-534.
[16] Kurosu T, Fujita M, Chiba K. Monitoring of rice crop growth from space using the ers-1 c-band sar[J]. Geoscience&Remote Sensing IEEE Transactions on Geoscience and Remote Sensing, 1995,33(4):1092-1096.
[17] Chen J, Lin H, Pei Z. Application of envisat asar data in mapping rice crop growth in southern china[J]. IEEE Geoscience&Remote Sensing Letters, 2007,4(3):431-435.
[18] Panigrahy S, Jain V, Patnaik C. Identification of rice crop in bangladesh using temporal c-band sar-a feasibility study[J]. Journal of the Indian Society of Remote Sensing, 2012,40(4):599-606.
[19] Prucker S, Meier W, Stricker W. Evaluation of radarsat standard beam data for identification of potato and rice crops in india[J]. Isprs Journal of Photogrammetry&Remote Sensing, 1999,54(4):254-262.
[20] Torres R, Snoeij P, Geudtner D, et al. Gmes sentinel-1 mission[J]. Remote Sensing of Environment, 2012,120:9-24.
[21]吴炳方,许文波,孙明,等.高精度作物分布图制作[J].遥感学报,2004,8(6):688-695.[Wu B F, Xu W B, Sun M, et al. Mapping Crop Distribution with High Accuracy[J].Journal of Remote Sensing, 2004,8(6):688-695.]
[22] Thenkabail P S, Biradar C M, Turral H, et al. An irrigated area map of the world(1999)derived from remote sensing[M]. Iwmi Books Reports, 2006,36:600-605.
[23] Mandanici E, Bitelli G. Preliminary comparison of sentinel-2 and landsat 8 imagery for a combined use[J]. Remote Sensing, 2016,8(12):1014.
[24] Gorelick N, Hancher M, Dixon M, et al. Google earth engine:Planetary-scale geospatial analysis for everyone[J].Remote Sensing of Environment 2017,202:18-27.
[25] Hansen M C, Potapov P V, Moore R, et al. High-resolution global maps of 21st-century forest cover change[J].Science, 2014,342(6160):850-853.
[26] Pekel J F, Cottam A, Gorelick N, et al. High-resolution mapping of global surface water and its long-term changes[J]. Nature, 2016,540(7633):418-422.
[27]王斌,陈小敏,钟曼茜,等.海南水稻生育期的时空变化特征及对气候变暖的响应[J].热带作物学报,2017,38(3):415-420.[Wang B, Chen X M, Zhong M X, et al. Spatialtemporal variation and response to climate warmer upon phenology of Hainan rice[J]. Chinese Journal of Tropical Crops, 2017,38(3):415-420.]
[28] Li Q, Wu B, Xu W. Accruacy assessment of crop type proportion using GVG instrument on transect line[J]. Journal of Remote Sensing, 2004,6:253.
[29]吴炳方,田亦陈,李强子. GVG农情采样系统及其应用[J].遥感学报,2004,8(6):570-580.[Wu B F, Tian Y C, Li Q Z. GVG a crop type proportion sampling instrumentgvg[J]. Journal of Remote Sensing, 2004,8(6):570-580.]
[30] Chander G, Markham B L, Helder D L. Summary of current radiometric calibration coefficients for landsat mss,tm, etm+, and eo-1 ali sensors[J]. Remote Sensing of Environment, 2009,113(5):893-903.
[31] Oreopoulos L, Wilson M J, Várnai T. Implementation on landsat data of a simple cloud-mask algorithm developed for modis land bands[J]. IEEE Geoscience&Remote Sensing Letters, 2011,8(4):597-601.
[32] The European Space Agency. Sentinel-2 user handbook[R]. European Space Agency(ESA), 2015:64.
[33] Farr T G, Kobrick M. The shuttle radar topography mission:A global dem[J]. American Geophysical Union,1999,45:37-55.
[34] Ouyang Z, Zheng H, Xiao Y, et al. Improvements in ecosystem services from investments in natural capital[J].Science, 2016,352(6292):1455-1459.
[35] Tan S, Wu B, Yan N, et al. An ndvi-based statistical et downscaling method[J]. Water, 2017,9(12):995.
[36] Tan S, Wu B, Yan, N, et al. Satellite-based water consumption dynamics monitoring in an extremely arid area[J]. Remote Sensing, 2018,10(9):1399.
[37] Huete A, Liu H, Batchily K, et al. A comparison of vegetation indices over a global set of tm images for eos-modis[J]. Remote Sensing of Environment, 1997,59(3):440-451.
[38] Gao B C. In Ndwi-a normalized difference water index for remote sensing of vegetation liquid water from space[J]. Imaging Spectrometry, 1995,58(3):257-266.
[39] Breiman L. Bagging predictors machine learning[J]. Machine Learning 1996,24(2):123-140.
[40] Ho T. The random subspace method for constructing decision forests[J]. IEEE Transactions on Pattern Analysis&Machine Intelligence, 1998,20(8):832-844.
[41] Pelletier C, Valero S, Inglada J, et al. Assessing the robustness of random forests to map land cover with high resolution satellite image time series over large areas[J]. Remote Sensing of Environment, 2016,187:156-168.
[42] Sharma R, Tateishi R, Hara K, et al. Production of the japan 30-m land cover map of 2013-2015 using a random forests-based feature optimization approach[J]. Remote Sensing, 2016,8(5):429.
[43] Pal, M. Random forest classifier for remote sensing classification[J]. International Journal of Remote Sensing,2005,26(1):217-222.
[44] Azzari G, Lobell D. Landsat-based classification in the cloud:An opportunity for a paradigm shift in land cover monitoring[J]. Remote Sensing of Environment, 2017,202:64-74.
[45]朱良,平博,苏奋振,等.多时相tm影像决策树模型的水稻识别提取[J].地球信息科学学报,2013,15(3):446-451.[Zhu L, Ping B, Su F Z, et al. Mapping paddy rice based on multi-temporal TM images and decision tree model[J].Journal of Geo-Information Science, 2013,15(3):446-451.]
[46]魏新彩,王新生,刘海,等. Hj卫星图像水稻种植面积的识别分析[J].地球信息科学学报,2012,14(3):382-388.[Wei X C, Wang X S, Liu H, et al. Mapping paddy rice based on HJ images[J]. Journal of Geo-Information Science, 2012,14(3):382-388.]
[47]陈燕丽,莫伟华,莫建飞,等.基于面向对象分类的南方水稻种植面积提取方法[J].遥感技术与应用,2011,26(2):163-168.[Chen Y L, Mo W H, Mo J F, et al. Mapping paddy rice in Southern China based on object-oriented method[J]. Remote Sensing Technology and Application,2011,26(2):163-168.]
[48]吴金胜,刘红利,张锦水.无人机遥感影像面向对象分类方法估算市域水稻面积[J].农业工程学报,2018,34(1):70-77.[Wu J S, Liu H L, Zhang J S. Mapping paddy rice in city based on UAV images and object-oriented method[J]. Transactions of the Chinese Society of Agricultural Engyneering, 2018,34(1):70-77.]
[2] Khush G S. What it will take to feed 5.0 billion rice consumers in 2030[J]. Plant Molecular Biology, 2005,59(1):1-6.
[3] Zhang Q, Zhang W, Li T, et al. Projective analysis of staple food crop productivity in adaptation to future climate change in china[J]. International Journal of Biometeorology, 2017,61(8):1-16.
[4] Bouman, B. How much water does rice use[J]? Rice Today, 2009,69:115-133.
[5] Tao F, Hayashi Y, Zhang, Z, Sakamoto T, et al. Global warming, rice production, and water use in china:Developing a probabilistic assessment[J]. Agricultural&Forest Meteorology, 2008,148(1):94-110.
[6] Knauer K, Knauer K. Remote sensing of rice crop areas[J]. International Journal for Remote Sensing, 2013,34(6):2101-2139.
[7] Ramankutty N, Foley J. Characterizing patterns of global land use:An analysis of global croplands data[J]. Global Biogeochemical Cycles, 1998,12(4):667-685.
[8] Ramankutty N, Foley J A. Estimating historical changes in global land cover:Croplands from 1700 to 1992[J].Global Biogeochemical Cycles, 1999,13(4):997-1027.
[9] Li P, Feng Z M, Jiang, et al. Changes in rice cropping systems in the poyang lake region, china during 2004-2010[J].Journal of Geographical Sciences, 2012,22(4):653-668.
[10] Thenkabail P S. Mapping rice areas of south asia using modis multitemporal data[J]. Journal of Applied Remote Sensing, 2011,5(1):863-871.
[11]黄青,吴文斌,邓辉,等.2009年江苏省冬小麦和水稻种植面积信息遥感提取及长势监测[J].江苏农业科学,2016(6):508-511.[Huang Q, Wu W B, Deng H, et al. Mapping area and condition monitoring of winter wheat and rice in JiangSu province in 2009[J]. Jiangsu Agricultural Sciences, 2016(6):508-511.]
[12] Singha M, Wu B, Zhang M. An object-based paddy rice classification using multi-spectral data and crop phenology in assam, northeast india[J]. Remote Sensing, 2016,8(6):479.
[13] Xiao X, Boles S, Liu J, et al. Mapping paddy rice agriculture in southern china using multi-temporal modis images[J]. Remote Sensing of Environment, 2005,95(4):480-492.
[14] Zhang X, Zhang M, Zheng Y, et al. Crop mapping using proba-v time series data at the yucheng and hongxing farm in china[J]. Remote Sensing, 2016,8(11):915.
[15] Choudhury I, Chakraborty M. Sar signature investigation of rice crop using radarsat data[J]. International Journal of Remote Sensing, 2006,27(3):519-534.
[16] Kurosu T, Fujita M, Chiba K. Monitoring of rice crop growth from space using the ers-1 c-band sar[J]. Geoscience&Remote Sensing IEEE Transactions on Geoscience and Remote Sensing, 1995,33(4):1092-1096.
[17] Chen J, Lin H, Pei Z. Application of envisat asar data in mapping rice crop growth in southern china[J]. IEEE Geoscience&Remote Sensing Letters, 2007,4(3):431-435.
[18] Panigrahy S, Jain V, Patnaik C. Identification of rice crop in bangladesh using temporal c-band sar-a feasibility study[J]. Journal of the Indian Society of Remote Sensing, 2012,40(4):599-606.
[19] Prucker S, Meier W, Stricker W. Evaluation of radarsat standard beam data for identification of potato and rice crops in india[J]. Isprs Journal of Photogrammetry&Remote Sensing, 1999,54(4):254-262.
[20] Torres R, Snoeij P, Geudtner D, et al. Gmes sentinel-1 mission[J]. Remote Sensing of Environment, 2012,120:9-24.
[21]吴炳方,许文波,孙明,等.高精度作物分布图制作[J].遥感学报,2004,8(6):688-695.[Wu B F, Xu W B, Sun M, et al. Mapping Crop Distribution with High Accuracy[J].Journal of Remote Sensing, 2004,8(6):688-695.]
[22] Thenkabail P S, Biradar C M, Turral H, et al. An irrigated area map of the world(1999)derived from remote sensing[M]. Iwmi Books Reports, 2006,36:600-605.
[23] Mandanici E, Bitelli G. Preliminary comparison of sentinel-2 and landsat 8 imagery for a combined use[J]. Remote Sensing, 2016,8(12):1014.
[24] Gorelick N, Hancher M, Dixon M, et al. Google earth engine:Planetary-scale geospatial analysis for everyone[J].Remote Sensing of Environment 2017,202:18-27.
[25] Hansen M C, Potapov P V, Moore R, et al. High-resolution global maps of 21st-century forest cover change[J].Science, 2014,342(6160):850-853.
[26] Pekel J F, Cottam A, Gorelick N, et al. High-resolution mapping of global surface water and its long-term changes[J]. Nature, 2016,540(7633):418-422.
[27]王斌,陈小敏,钟曼茜,等.海南水稻生育期的时空变化特征及对气候变暖的响应[J].热带作物学报,2017,38(3):415-420.[Wang B, Chen X M, Zhong M X, et al. Spatialtemporal variation and response to climate warmer upon phenology of Hainan rice[J]. Chinese Journal of Tropical Crops, 2017,38(3):415-420.]
[28] Li Q, Wu B, Xu W. Accruacy assessment of crop type proportion using GVG instrument on transect line[J]. Journal of Remote Sensing, 2004,6:253.
[29]吴炳方,田亦陈,李强子. GVG农情采样系统及其应用[J].遥感学报,2004,8(6):570-580.[Wu B F, Tian Y C, Li Q Z. GVG a crop type proportion sampling instrumentgvg[J]. Journal of Remote Sensing, 2004,8(6):570-580.]
[30] Chander G, Markham B L, Helder D L. Summary of current radiometric calibration coefficients for landsat mss,tm, etm+, and eo-1 ali sensors[J]. Remote Sensing of Environment, 2009,113(5):893-903.
[31] Oreopoulos L, Wilson M J, Várnai T. Implementation on landsat data of a simple cloud-mask algorithm developed for modis land bands[J]. IEEE Geoscience&Remote Sensing Letters, 2011,8(4):597-601.
[32] The European Space Agency. Sentinel-2 user handbook[R]. European Space Agency(ESA), 2015:64.
[33] Farr T G, Kobrick M. The shuttle radar topography mission:A global dem[J]. American Geophysical Union,1999,45:37-55.
[34] Ouyang Z, Zheng H, Xiao Y, et al. Improvements in ecosystem services from investments in natural capital[J].Science, 2016,352(6292):1455-1459.
[35] Tan S, Wu B, Yan N, et al. An ndvi-based statistical et downscaling method[J]. Water, 2017,9(12):995.
[36] Tan S, Wu B, Yan, N, et al. Satellite-based water consumption dynamics monitoring in an extremely arid area[J]. Remote Sensing, 2018,10(9):1399.
[37] Huete A, Liu H, Batchily K, et al. A comparison of vegetation indices over a global set of tm images for eos-modis[J]. Remote Sensing of Environment, 1997,59(3):440-451.
[38] Gao B C. In Ndwi-a normalized difference water index for remote sensing of vegetation liquid water from space[J]. Imaging Spectrometry, 1995,58(3):257-266.
[39] Breiman L. Bagging predictors machine learning[J]. Machine Learning 1996,24(2):123-140.
[40] Ho T. The random subspace method for constructing decision forests[J]. IEEE Transactions on Pattern Analysis&Machine Intelligence, 1998,20(8):832-844.
[41] Pelletier C, Valero S, Inglada J, et al. Assessing the robustness of random forests to map land cover with high resolution satellite image time series over large areas[J]. Remote Sensing of Environment, 2016,187:156-168.
[42] Sharma R, Tateishi R, Hara K, et al. Production of the japan 30-m land cover map of 2013-2015 using a random forests-based feature optimization approach[J]. Remote Sensing, 2016,8(5):429.
[43] Pal, M. Random forest classifier for remote sensing classification[J]. International Journal of Remote Sensing,2005,26(1):217-222.
[44] Azzari G, Lobell D. Landsat-based classification in the cloud:An opportunity for a paradigm shift in land cover monitoring[J]. Remote Sensing of Environment, 2017,202:64-74.
[45]朱良,平博,苏奋振,等.多时相tm影像决策树模型的水稻识别提取[J].地球信息科学学报,2013,15(3):446-451.[Zhu L, Ping B, Su F Z, et al. Mapping paddy rice based on multi-temporal TM images and decision tree model[J].Journal of Geo-Information Science, 2013,15(3):446-451.]
[46]魏新彩,王新生,刘海,等. Hj卫星图像水稻种植面积的识别分析[J].地球信息科学学报,2012,14(3):382-388.[Wei X C, Wang X S, Liu H, et al. Mapping paddy rice based on HJ images[J]. Journal of Geo-Information Science, 2012,14(3):382-388.]
[47]陈燕丽,莫伟华,莫建飞,等.基于面向对象分类的南方水稻种植面积提取方法[J].遥感技术与应用,2011,26(2):163-168.[Chen Y L, Mo W H, Mo J F, et al. Mapping paddy rice in Southern China based on object-oriented method[J]. Remote Sensing Technology and Application,2011,26(2):163-168.]
[48]吴金胜,刘红利,张锦水.无人机遥感影像面向对象分类方法估算市域水稻面积[J].农业工程学报,2018,34(1):70-77.[Wu J S, Liu H L, Zhang J S. Mapping paddy rice in city based on UAV images and object-oriented method[J]. Transactions of the Chinese Society of Agricultural Engyneering, 2018,34(1):70-77.]