基于多层神经网络与Sentinel-2数据的大豆种植区识别方法
|
摘要
大豆作为全球最重要的油料作物,是中国进口的大宗农产品,对其种植区的精准识别是决策制定、种植结构调整基础,对国家粮食安全有重要意义。本文利用Sentinel-2作为数据源,利用多层神经网络方法与对大豆进行提取,并与随机森林、决策树、支持向量机等机器学习进行对比,发现F1-Socre指标显示多层神经网络的分类精度最高,为93.53%,其次为随机森林、支持向量机、决策树。将神经网络分类结果与SLIC面向对象分割聚合之后,结果既忽略了同一地块的微小差别,又区分出了不同地块的作物差异,很好的体现了大豆的分布。Sentinel-2数据是进行大尺度大豆种植监测的绝佳数据源,大豆与玉米等其他作物在第二个红边波段的反射率有较为明显的差异。多层神经网络方法在图像分类任务中表现出色,结合图像分割算法精度可达到95.51%,可以满足大豆种植面积监测的需求。
As the most important oil crop in the world, soybean is a large-scale agricultural product that China imports. The accurate identification of its planting area is the basis for decision-making and planting structure adjustment, and is of great significance to national food security. Sentinel-2 was used as a data source and multilayer neural network was employed to map soybean cropped area. Besides, visible and infrared bands, three rededge bands were also selected after radiation and atmospheric correction using the Sentinel-2 Toolbox.According to our test, 8-hidden-layer conducted using Scikit-learn package in Python2.7 was the optimal structure for identifying soybean and other crops. Simple linear iterative clustering(SLIC), the state-of-art segmentation algorithm, was performed to segment the remote sensed image. This method combined fivedimensional color and the image plane space to efficiently generate compact and nearly uniform super pixels. To remove the"salt and pepper effect", the pixel-based result was integrated with the object output from the SLIC.If the pixel as soybean in an object accounted for less than 50%, this object was eliminated in the fusion map.The results showed that the overall accuracy of multi-layer neural network was 93.95%, which was highest and followed by the support vector machine, decision tree, and random forest. Then, the neural network classification was selected as the best result to integrate with SLIC object-oriented segmentation, and the results ignored the small differences of the same land and distinguish the crop differences of different blocks compared with the segmentation in eCognition software. Sentinel-2 data is an appropriate data source for large-scale soybean planting mapping. According to feature importance derived from the random forest classifier, near-infrared band is the most critical feature for classification, followed by third red edge band(Band 7), fourth red edge 4 band(Band 8), red band, and second red edge band(Band 6). The reflectance values of soybeans and other crops in the second red edge band were different, indicating a huge potential in crop type identifying. In the future, the red edge band can be introduced more into crop type even landscape classification. The multi-layer neural network method performs well in the image classification task and had similar or better overall accuracy value compared with other outstanding machine learning classifier including SVM, decision tree, and random forest.Combined with the image segmentation algorithm, such as SLIC, multi-layer neural network can map soybean cropped area with an accuracy high up to 95.51%, which can serve for soybean planting area monitoring in a large area.
As the most important oil crop in the world, soybean is a large-scale agricultural product that China imports. The accurate identification of its planting area is the basis for decision-making and planting structure adjustment, and is of great significance to national food security. Sentinel-2 was used as a data source and multilayer neural network was employed to map soybean cropped area. Besides, visible and infrared bands, three rededge bands were also selected after radiation and atmospheric correction using the Sentinel-2 Toolbox.According to our test, 8-hidden-layer conducted using Scikit-learn package in Python2.7 was the optimal structure for identifying soybean and other crops. Simple linear iterative clustering(SLIC), the state-of-art segmentation algorithm, was performed to segment the remote sensed image. This method combined fivedimensional color and the image plane space to efficiently generate compact and nearly uniform super pixels. To remove the"salt and pepper effect", the pixel-based result was integrated with the object output from the SLIC.If the pixel as soybean in an object accounted for less than 50%, this object was eliminated in the fusion map.The results showed that the overall accuracy of multi-layer neural network was 93.95%, which was highest and followed by the support vector machine, decision tree, and random forest. Then, the neural network classification was selected as the best result to integrate with SLIC object-oriented segmentation, and the results ignored the small differences of the same land and distinguish the crop differences of different blocks compared with the segmentation in eCognition software. Sentinel-2 data is an appropriate data source for large-scale soybean planting mapping. According to feature importance derived from the random forest classifier, near-infrared band is the most critical feature for classification, followed by third red edge band(Band 7), fourth red edge 4 band(Band 8), red band, and second red edge band(Band 6). The reflectance values of soybeans and other crops in the second red edge band were different, indicating a huge potential in crop type identifying. In the future, the red edge band can be introduced more into crop type even landscape classification. The multi-layer neural network method performs well in the image classification task and had similar or better overall accuracy value compared with other outstanding machine learning classifier including SVM, decision tree, and random forest.Combined with the image segmentation algorithm, such as SLIC, multi-layer neural network can map soybean cropped area with an accuracy high up to 95.51%, which can serve for soybean planting area monitoring in a large area.
引文
[1]国家统计局.中华人民共和国2017年国民经济和社会发展统计公报[P]. 2018. http://www.stats.gov.cn/tjsj/.[National statistic bureau. Statistical bulletin on national economic and social development in 2017 of People's Republic of China[P]. 2018, http://www.stats.gov.cn/tjsj/.]
[2] Zhang J H, Feng L L, Yao F M. Improved maize cultivated area estimation over a large scale combining MODISEVI time series data and crop phenological information[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2014,94:102-113.
[3] Zhong L H, Grong P, Biging G S. Efficient corn and soybean mapping with temporal extendability:A multi-year experiment using Landsat imagery[J]. Remote Sensing of Environment, 2014,140:1-13.
[4] Song X P, Potapov P V, Krylov A, et al. National-scale soybean mapping and area estimation in the United States using medium resolution satellite imagery and field survey[J]. Remote Sensing of Environment, 2017,190:383-395.
[5]吴炳方,范锦龙,田亦陈等.全国作物种植结构快速调查技术与应用[J].遥感学报,2004,8(6):618-627.[Wu B F,Fan J L, Tian Y C, et al. A method for crop planting structure inventory and its application[J]. Journal of Remote Sensing, 2004,8(6):618-627.]
[6]刘佳,王利民,滕飞等.RapidEye卫星红边波段对农作物面积提取精度的影响[J].农业工程学报,2016,32(13):140-148.[Liu J, Wang L M, Teng F, et al. Impact of rededge waveband of RapidEye satellite on estimation accuracy of crop planting area[J]. Transactions of the Chinese Society of Agricultural Engineering, 2016,32(13):140-148.]
[7] Zeng L L, Wardlow B D, Wang R, et al. A hybrid approach for detecting corn and soybean phenology with time-series MODIS data[J]. Remote sensing of environment, 2016,181:237-250.
[8]黄健熙,侯矞焯,苏伟等.基于GF-1 WFV数据的玉米与大豆种植面积提取方法[J].农业工程学报,2017,33(7):164-170.[Huang J X, Hou Y Z, Su W, et al. Mapping corn and soybean cropped area with GF-1 WFV data[J]. Transactions of the Chinese Society of Agricultural Engineering, 2017,33(7):164-170.]
[9] ZHANG X, WU B F, PONCE-CAMPOS G, et al. Mapping up-to-Date Paddy Rice Extent at 10 M Resolution in China through the Integration of Optical and Synthetic Aperture Radar Images[J]. Remote Sensing, 2018,10(8):1200.
[10] Belgiu M, Dr?gu?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.
[11] Zheng B J, Myint S W, Thenkabail P S, et al. A support vector machine to identify irrigated crop types using timeseries Landsat NDVI data[J]. International Journal of Applied Earth Observation and Geoinformation, 2015,34:103-112.
[12] Hecht-nielsen R. Theory of the backpropagation neural network[M]. Washington:Neural networks for perception. Elsevier. 1992:65-93.
[13] Mas J F, Flores J J. The application of artificial neural networks to the analysis of remotely sensed data[J]. International Journal of Remote Sensing, 2008,29(3):617-663.
[14] Glorot X, Bordes A, Bengio Y. Deep sparse rectifier neural networks[C]. In Proceedings of the fourteenth international conference on artificial intelligence and statistics,2011:315-323.
[15]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 instrument[J]. Journal of Remote Sensing, 2004,8(6):570-580.]
[16] Achanta R, Shaji A, Smith K, et al. SLIC superpixels compared to state-of-the-art superpixel methods[J]. IEEE transactions on pattern analysis and machine intelligence,2012,34(11):2274-2282.
[17] Bargiel D. A new method for crop classification combining time series of radar images and crop phenology information[J]. Remote Sensing of Environment, 2017,198:369-383.
[18]徐晗泽宇,刘冲,王军邦,等.Google Earth Engine平台支持下的赣南柑橘果园遥感提取研究[J].地球信息科学学报,2018,20(3):396-404.[Xu H Z Y, Liu C, Wang J B, et al. Study on extraction of citrus orchard in Gannan region based on google earth engine platform[J]. Journal of Geoinformation Science, 2018,20(3):396-404.]
[19] Gallego F J, Kussul N, Skakun S, et al. Efficiency assessment of using satellite data for crop area estimation in Ukraine[J]. International Journal of Applied Earth Observation and Geoinformation, 2014,29:22-30.
[20] Pe?a-barragán J M, Ngugi M K, Plant R E, et al. Objectbased crop identification using multiple vegetation indices, textural features and crop phenology[J]. Remote Sensing of Environment, 2011,115(6):1301-1316.
[2] Zhang J H, Feng L L, Yao F M. Improved maize cultivated area estimation over a large scale combining MODISEVI time series data and crop phenological information[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2014,94:102-113.
[3] Zhong L H, Grong P, Biging G S. Efficient corn and soybean mapping with temporal extendability:A multi-year experiment using Landsat imagery[J]. Remote Sensing of Environment, 2014,140:1-13.
[4] Song X P, Potapov P V, Krylov A, et al. National-scale soybean mapping and area estimation in the United States using medium resolution satellite imagery and field survey[J]. Remote Sensing of Environment, 2017,190:383-395.
[5]吴炳方,范锦龙,田亦陈等.全国作物种植结构快速调查技术与应用[J].遥感学报,2004,8(6):618-627.[Wu B F,Fan J L, Tian Y C, et al. A method for crop planting structure inventory and its application[J]. Journal of Remote Sensing, 2004,8(6):618-627.]
[6]刘佳,王利民,滕飞等.RapidEye卫星红边波段对农作物面积提取精度的影响[J].农业工程学报,2016,32(13):140-148.[Liu J, Wang L M, Teng F, et al. Impact of rededge waveband of RapidEye satellite on estimation accuracy of crop planting area[J]. Transactions of the Chinese Society of Agricultural Engineering, 2016,32(13):140-148.]
[7] Zeng L L, Wardlow B D, Wang R, et al. A hybrid approach for detecting corn and soybean phenology with time-series MODIS data[J]. Remote sensing of environment, 2016,181:237-250.
[8]黄健熙,侯矞焯,苏伟等.基于GF-1 WFV数据的玉米与大豆种植面积提取方法[J].农业工程学报,2017,33(7):164-170.[Huang J X, Hou Y Z, Su W, et al. Mapping corn and soybean cropped area with GF-1 WFV data[J]. Transactions of the Chinese Society of Agricultural Engineering, 2017,33(7):164-170.]
[9] ZHANG X, WU B F, PONCE-CAMPOS G, et al. Mapping up-to-Date Paddy Rice Extent at 10 M Resolution in China through the Integration of Optical and Synthetic Aperture Radar Images[J]. Remote Sensing, 2018,10(8):1200.
[10] Belgiu M, Dr?gu?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.
[11] Zheng B J, Myint S W, Thenkabail P S, et al. A support vector machine to identify irrigated crop types using timeseries Landsat NDVI data[J]. International Journal of Applied Earth Observation and Geoinformation, 2015,34:103-112.
[12] Hecht-nielsen R. Theory of the backpropagation neural network[M]. Washington:Neural networks for perception. Elsevier. 1992:65-93.
[13] Mas J F, Flores J J. The application of artificial neural networks to the analysis of remotely sensed data[J]. International Journal of Remote Sensing, 2008,29(3):617-663.
[14] Glorot X, Bordes A, Bengio Y. Deep sparse rectifier neural networks[C]. In Proceedings of the fourteenth international conference on artificial intelligence and statistics,2011:315-323.
[15]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 instrument[J]. Journal of Remote Sensing, 2004,8(6):570-580.]
[16] Achanta R, Shaji A, Smith K, et al. SLIC superpixels compared to state-of-the-art superpixel methods[J]. IEEE transactions on pattern analysis and machine intelligence,2012,34(11):2274-2282.
[17] Bargiel D. A new method for crop classification combining time series of radar images and crop phenology information[J]. Remote Sensing of Environment, 2017,198:369-383.
[18]徐晗泽宇,刘冲,王军邦,等.Google Earth Engine平台支持下的赣南柑橘果园遥感提取研究[J].地球信息科学学报,2018,20(3):396-404.[Xu H Z Y, Liu C, Wang J B, et al. Study on extraction of citrus orchard in Gannan region based on google earth engine platform[J]. Journal of Geoinformation Science, 2018,20(3):396-404.]
[19] Gallego F J, Kussul N, Skakun S, et al. Efficiency assessment of using satellite data for crop area estimation in Ukraine[J]. International Journal of Applied Earth Observation and Geoinformation, 2014,29:22-30.
[20] Pe?a-barragán J M, Ngugi M K, Plant R E, et al. Objectbased crop identification using multiple vegetation indices, textural features and crop phenology[J]. Remote Sensing of Environment, 2011,115(6):1301-1316.