基于人工神经网络多源数据融合的子像元冬油菜提取——以两湖平原为例

doi:10.31497/zrzyxb.20190514

摘要/Abstract

摘要：

油菜是我国第五大农作物和重要的油料作物。获取油菜的种植分布信息对食用油市场的发展和粮食安全具有重要意义。两湖平原泛指包括湖北江汉平原和湖南洞庭湖平原在内的广大平原区域,是我国重要的粮棉油生产基地,“湖广熟,天下足”指的就是这一地区。由于耕地破碎,种植结构复杂,两湖平原轮作和间作的现象非常普遍,传统的遥感监测方法难以准确地获取冬油菜的空间分布。本文提出了一种基于人工神经网络ANN的子像元冬油菜提取方法,将时间序列MODIS-EVI和GF-1数据结合以提取两湖平原的冬油菜丰度信息。首先采用顺序前向选择SFS算法从时间序列MODIS-EVI数据集中进行物候特征优选;然后构建融合多源数据的ANN模型估算两湖平原的冬油菜丰度。结果表明：基于ANN方法获取的冬油菜分布具有较高的精度（ANN估算结果与GF-1和统计数据的验证精度分别为91.54%和74.70%）,在利用中分辨率影像进行大尺度冬油菜精细制图方面显示出巨大潜力,可为我国冬油菜的空间分布制图和时空格局分析提供技术方法。

关键词: 人工神经网络, 冬油菜, 子像元, 时间序列MODIS, 两湖平原

Abstract:

Rape is the fifth largest crop type and an important oil crop in China. Obtaining the distribution information of rape is of great significance to the development of edible oil market and food security. Jianghan and Dongting Lake Plain are important production bases of grain, cotton and edible oil in China. The crop rotation and intercropping are very common due to the fragmented cropland fields and diversity of landscape. Traditional remote sensing monitoring methods are difficult to assess the spatial distribution and interannual variation of planting areas. In this paper, an Artificial Neural Network-based method for extracting winter rape from sub-pixels was proposed, and the MODIS and GF-1 high resolution data were combined to obtain winter rape abundance on the JPDLP. First, the Sequential Forward Selection algorithm was used to select phenological feature from the MODIS time series dataset. Then an ANN model fusing multi-source data was built to estimate winter rape abundance. The results showed that the distribution information of winter rape obtained by this method has high accuracy (the extraction accuracies of ANN modeled versus GF-1 and census data were 91.54% and 74.70%, respectively). This method showed that there is a great potential in the large-scale winter rape mapping using coarse resolution images. The results of this paper can provide technical methods for the spatial pattern evolution and spatial-temporal dynamics analysis of winter rape in China.

Key words: Artificial Neural Network, winter rape, sub-pixel, MODIS time series, Jianghan and Dongting Lake Plain

刘文斌, 陶建斌, 徐猛, 陈瑞卿, 郭洋. 基于人工神经网络多源数据融合的子像元冬油菜提取——以两湖平原为例[J]. 自然资源学报, 2019, 34(5): 1079-1092.

Wen-bin LIU, Jian-bin TAO, Meng XU, Rui-qing CHEN, Yang GUO. A study of winter rape extraction at sub-pixel fusing multi-source data based on Artificial Neural Networks:A case study of Jianghan and Dongting Lake Plain[J]. JOURNAL OF NATURAL RESOURCES, 2019, 34(5): 1079-1092.

图/表 16

图1

表1

表2

图2

图3

图4

图5

图6

图7

图8

表3

图9

图10

图11

图12

图13

参考文献 36

[1]	FAS. USDA. GOV."Oilseeds: World markets and trade". , 2018.
[2]	TAO J B, WU W B, LIU W B.Spatial-temporal dynamics of cropping frequency in Hubei province over 2001-2015. Sensors, 2017, 17(11): 2622.
[3]	LIU J Y, KUANG W H, ZHANG Z X, et al.Spatiotemporal characteristics, patterns, and causes of land-use changes in China since the late 1980s. Journal of Geographical Sciences, 2014, 24(2): 195-210.
[4]	李燕玲, 刘爱民. 长江流域冬季农业主要作物的耕地竞争机制及案例研究. 长江流域资源与环境, 2009, 18(2): 146-151.
	[LI Y L, LIU A M.Competition mechanism of cultivated landresources in winter agriculture in the Yangtze Basin: Theory and case study. Resources & Environment in the Yangtze Basin, 2009, 18(2): 146-151.]
[5]	佘宝, 黄敬峰, 石晶晶, 等. 基于红边位置变化特征的油菜种植区域提取. 农业工程学报, 2013, 29(15): 145-152.
	[SHE B, HUANG J F, SHI J J, et al.Extracting oilseed rape growing regions based on variation characteristics of red edge position. Transactions of the CSAE, 2013, 29(15): 145-152.]
[6]	PAN Y Z, LI L, ZHANG J S, et al.Winter wheat area estimation from MODIS-EVI time series data using the Crop Proportion Phenology Index. Remote Sensing of Environment, 2012, 119(3): 232-242.
[7]	ESTEL S, KUEMMERLE T, ALCÁNTARA C, et al. Mapping farmland abandonment and recultivation across Europe using MODIS NDVI time series. Remote Sensing of Environment, 2015, 163: 312-325.
[8]	PAN Z K, HUANG J F, ZHOU Q B, et al.Mapping crop phenology using NDVI time-series derived from HJ-1 A/B data. International Journal of Applied Earth Observation & Geoinformation, 2015, 34(1): 188-197.
[9]	QIU B W, LI W J, TANG Z H, et al.Mapping paddy rice areas based on vegetation phenology and surface moisture conditions. Ecological Indicators, 2015, 56: 79-86.
[10]	平跃鹏, 臧淑英. 基于MODIS时间序列及物候特征的农作物分类. 自然资源学报, 2016, 31(3): 503-513.
	[PING Y P, ZANG S Y.Crop identification based on MODIS NDVI time-series data and phenological characteristics. Journal of Natural Resources, 2016, 31(3): 503-513.]
[11]	TAO J B, WU W B, ZHOU Y, et al.Mapping winter wheat using phenological feature of peak before winter on the North China Plain based on time-series MODIS data. Journal of Integrative Agriculture, 2017, 16(2): 348-359.
[12]	QIU B W, LUO Y H, TANG Z H, et al.Winter wheat mapping combining variations before and after estimated heading dates. Isprs Journal of Photogrammetry & Remote Sensing, 2017, 123: 35-46.
[13]	HUANG J X, TIAN L Y, LIANG S L, et al.Improving winter wheat yield estimation by assimilation of the leaf area index from Landsat TM and MODIS data into the WOFOST model. Agricultural & Forest Meteorology, 2015, 204: 106-121.
[14]	王凯, 张佳华. 基于MODIS数据的湖北省油菜种植分布信息提取. 国土资源遥感, 2015, 27(3): 65-70.
	[WANG K, ZHANG J H.Extraction of rape seed cropping distribution information in Hubei province based on modis images. Remote Sensing for Land and Resources, 2015, 27(3): 65-70.]
[15]	尤慧, 苏荣瑞, 肖玮钰, 等. 基于MODISEVI时序数据的江汉平原油菜种植分布信息提取. 国土资源遥感, 2018, 30(1): 173-179.
	[YOU H, SU R R, XIAO W Y, et al.Extraction of rape planting distribution information in Jianghan Plain based on MODIS EVI time series data. Remote Sensing for Land and Resources, 2018, 30(1): 173-179.]
[16]	王松林, 张佳华, 刘学锋. 基于MODIS多时相的江苏启东市油菜种植面积提取. 遥感技术与应用, 2015, 30(5): 946-951.
	[WANG S L, ZHANG J H, LIU X F.Extraction of rape planting areas based on multi-temporal MODIS data in Qidong, Jiangsu province. Remote Sensing Technology & Application, 2015, 30(5): 946-951.]
[17]	王立辉, 黄进良, 孙俊英. 基于线性光谱混合模型的油菜种植面积遥感监测方法研究. 遥感信息, 2010, 2010(3): 55-59.
	[WANG L H, HUANG J L, SUN J Y.Study on remote sensing monitoring of rape acreage based on Linear Spectral Mixture Model. Remote Sensing Information, 2010, 32(3): 55-59.]
[18]	张煦, 马驿, 郑雯, 等. 基于时序MODIS-NDVI的油菜种植面积变化趋势分析: 以江汉平原为例. 长江流域资源与环境, 2016, 25(3): 412-419.
	[ZHANG X, MA Y, ZHENG W, et al.Variation trend of rape cultivation area based on MODIS-NDVI time series data: A case in Jianghan Plain. Resources & Environment in the Yangtze Basin, 2016, 25(3): 412-419.]
[19]	DOBREVA I D, KLEIN A G.Fractional snow cover mapping through artificial neural network analysis of MODIS surface reflectance. Remote Sensing of Environment, 2011, 115(12): 3355-3366.
[20]	CHEN C F, CHEN C R, SON N T, et al.Delineating rice cropping activities from MODIS data using wavelet transform and artificial neural networks in the Lower Mekong countries. Agriculture Ecosystems & Environment, 2012, 162(162): 127-137.
[21]	ATZBERGER C, REMBOLD F.Mapping the spatial distribution of winter crops at sub-pixel level using AVHRR NDVI time series and neural nets. Remote Sensing, 2013, 5(3): 1335-1354.
[22]	YANG S X, FENG Q S, LIANG T G, et al.Modeling grassland above-ground biomass based on artificial neural network and remote sensing in the Three-River Headwaters region. Remote Sensing of Environment, 2017, 204: 448-455.
[23]	VERBEIREN S, EERENS H, PICCARD I, et al.Sub-pixel classification of SPOT-VEGETATION time series for the assessment of regional crop areas in Belgium. International Journal of Applied Earth Observation & Geoinformation, 2008, 10(4): 486-497.
[24]	国家统计局. 中国统计年鉴. 北京: 中国统计出版社, 2016.
	[National Bureau of Statistics. China Statistical Yearbook 2016. Beijing: China Statistical Yearbook, 2016.]
[25]	湖北省统计局. 湖北农村统计年鉴. 北京: 中国统计出版社, 2016.
	[Hubei Provincial Bureau of Statistics. Hubei Rural Statistical Yearbook 2016. Beijing: Hubei Rural Statistical Yearbook, 2016.]
[26]	湖南省统计局.湖南农村统计年鉴. 北京: 中国统计出版社, 2016.
	[Hunan Provincial Bureau of Statistics. Hunan Rural Statistical Yearbook 2016. Beijing: Hunan Rural Statistical Yearbook, 2016.]
[27]	HUETE A, DIDAN K, MIURA T, et al.Overview of the radiometric and biophysical performance of the MODIS vegetation indices. Remote Sensing of Environment, 2015, 83(1): 195-213.
[28]	JÖNSSON P, EKLUNDH L. TIMESAT: A program for analyzing time-series of satellite sensor data. Computers & Geosciences, 2004, 30(8): 833-845.
[29]	JÖNSSON P, EKLUNDH L. Seasonality extraction by function fitting to time-series of satellite sensor data. IEEE Transactions on Geoscience & Remote Sensing, 2002, 40(8): 1824-1832.
[30]	VERVERIDIS D, KOTROPOULOS C.Fast and accurate sequential floating forward feature selection with the Bayes classifier applied to speech emotion recognition. Signal Processing, 2008, 88(12): 2956-2970.
[31]	MARCANO-CEDEÑO A, QUINTANILLA-DOMINGUEZ J, CORTINA-JANUCHS M G, et al. Feature selection using Sequential Forward Selection and classification applying Artificial Metaplasticity Neural Network. IECON 2010: 36th Conference on IEEE Industrial Electronics Society. IEEE, 2010: 2845-2850. Doi: 10.1109/IECON.2010.5675075.
[32]	杨福芹, 冯海宽, 李振海, 等. 基于赤池信息量准则的冬小麦叶面积指数高光谱估测. 农业工程学报, 2016, 32(3): 163-168.
	[YANG F Q, FENG H K, LI Z H, et al.Hyperspectral estimation of leaf area index for winter wheat based on Akaike's information criterion. Transactions of the CSAE, 2016, 32(3): 163-168.]
[33]	FRATE F D, PACIFICI F, SCHIAVON G, et al.Use of neural networks for automatic classification from high-resolution images. IEEE Transactions on Geoscience & Remote Sensing, 2007, 45(4): 800-809.
[34]	HECHT-NIELSEN R.Kolmogorov's Mapping Neural Network Existence Theorem. IEEE International Conference on Neural Networks. Piscataway: IEEE Press, 1987.
[35]	焦险峰, 杨邦杰, 裴志远. 基于分层抽样的中国水稻种植面积遥感调查方法研究. 农业工程学报, 2006, 22(5): 105-110.
	[JIAO X F, YANG B J, PEI Z Y.Paddy rice area estimation using a stratified sampling method with remote sensing in China. Transactions of the CSAE, 2006, 22(5): 105-110.]
[36]	彭代亮, 周炼清, 黄敬峰, 等. 基于抽样调查地块实测数据的省级水稻单产遥感估算. 农业工程学报, 2011, 27(9): 106-114.
	[PENG D L, ZHOU L Q, HUANG J F, et al.Rice yield estimation based on MODIS EVI and measured data derived from statistical sampling plots at province level. Transactions of the CSAE, 2011, 27(9): 106-114.]

数据类型	日期	空间分辨率/m	云覆盖/%
MOD13Q1	2014年10月至2015年6月	250	—
GF-1 WFV3	2015年3月21日	16	1
GL30-2010	2010年	30	—
Google Earth	2015.3.25	1	/

	10月			11月			12月			次年1月			2月			3月			4月			5月			6月
	上	中	下	上	中	下	上	中	下	上	中	下	上	中	下	上	中	下	上	中	下	上	中	下	上	中
儒略日	273~289			305~321			337~353			001~017			033~049			065~081			097~113			129~145			161
冬油菜	播种		育苗期						开盘期			蕾苔期					开花期			角果期		成熟期		收获
冬小麦		播种			育苗期		分蘖期								拔节期					抽穗期			成熟期		收获

类别	Google Earth
类别	冬油菜	其他	总数	用户精度/%
冬油菜	314	23	337	98.88
其他	6	532	538	93.18
总数	320	555	875	—
生产者精度/%	98.13	95.86	—	—