摘要
【目的】快速、准确地获得塑料大棚覆盖面积、位置分布等地理信息,对政府统筹规划、农业产值估算、环境资源保护等方面提供技术支持。【方法】利用无人机高分辨率遥感数据,以云南省昆明市呈贡区可乐村塑料大棚区为研究区域,分别采用基于像元和面向对象的方法,对研究区进行分割、分类、面积提取,结合研究区背景资料和实地调研结果,比较两种方法的精度,并计算了该研究区塑料大棚的面积。【结果】塑料大棚面积的面向对象提取总体精度为94.6%,Kappa系数为0.9133;而基于像元监督分类方法提取总体精度为68.3%,Kappa系数为0.5848。面向对象方法提取的面积与实地测量提取面积相比仍然存在一定的误差,在总面积上的差值为3.925 m~2,累积差值为22.475 m~2。若按差值比例来计算,大棚的估计面积为(26.371 42万±410.737)m~2,满足精度要求。【结论】面向对象方法精度远远高于基于像元方法的精度,且能有效抑制"椒盐现象",极大降低"同物异谱"、"同谱异物"的影响,提高了分类的精度。该文研究成果为无人机遥感在塑料大棚识别应用提供了参考。
[Purpose]It is of great significance for the government to make overall plans,estimate the agricultural output value,and protect the environmental resources,by quickly and accurately obtaining the geographical information of the covering area and location distribution of plastic greenhouse. [Method]With the rapid development of UAV remote sensing,the details contained in remote sensing images are increasingly rich. In this paper,we used UAV high-resolution remote sensing data,which was taken in the plastic greenhouse zone of Kele Village,Chenggong District of Kunming,Yunnan Province,to split,classify,and extract the plastic greenhouse area. The pixel-oriented method and object-oriented method were used in our study. Finally we compared the accuracy of the two methods and calculated the area of the plastic greenhouse in the study area. [Result]The results show that the overall accuracy of object extraction is 94.6% with a Kappa value of 0.9133,the overall accuracy of pixel extraction is about 68.3% with a Kappa value of 0.5848. The area extracted by the object-oriented method still has some error comparing to the measured area in the field. The difference in the total area is 3.925 m~2,and the cumulative difference is 22.475 m~2. The estimated area of the greenhouse is 263 714.2±410.737 m~2 according to the difference ratio,which meets the accuracy requirements. [Conclusion]The accuracy of object-oriented method is much higher than that of the pixel-oriented method. Object-oriented method can effectively reduce the "salt and pepper phenomenon",which greatly reduces the "synonyms spectrum" and "foreign body in the same spectrum",and finally improves the classification accuracy. Our results could provide a guidance for the methods of UAV remote sensing identification in plastic greenhouse.
引文
[1]Espi E,Salmeron A,Fontecha A,et al.Plastic films for agricultural applications.Journal of Plastic Film&Sheeting,2006,22(2):85~102.
[2]Tarantino E,Figorito B.Mapping rural areas with widespread plastic covered vineyards using true color aerial data.Remote Sensing,2012,4(7):1913~1928.
[3]Stiles M R.Templates for Modeling Temperature Response of Passive Greenhouses.Energy Engineering,2019,116(3):7~25.
[4]Wenqing H,Changrong Y,Shuang L.The use of plastic mulch film in typical cotton planting regions and the associated environmental pollution.Journal of Agro-Environment Science,2009,28(8):1618~1622.
[5]严昌荣,梅旭荣,何文清,等.农用地膜残留污染的现状与防治.农业工程学报,2006,22(11):269~272.
[6]黎明锋,杨文刚,阮仕明.塑料大棚小气候变化特征及其与蔬菜种植的关系.暴雨灾害,2004,23(4):27~29.
[7]王志盼,张清凌,钱静,等.基于增强型水体指数的大棚遥感检测研究--以广东江门地区为例.集成技术,2017,6(2):11~21.
[8]Yao Y,Wang S.Evaluating the effects of image texture analysis on plastic greenhouse segments via recognition of the OSI-USI-ETA-CEI pattern.Remote Sensing,2019,11(3):231.
[9]Lu L,Hang D,Di L.Threshold model for detecting transparent plastic-mulched landcover using moderate-resolution imaging spectroradiometer time series data:a case study in southern Xinjiang,China.Journal of Applied Remote Sensing,2015,9(1):974~982.
[10]Novelli A,Tarantino E.Combining ad hoc spectral indices based on Landsat-8 OLI/TIRS sensor data for the detection of plastic cover vineyard.Remote Sensing Letters,2015,6(12):933~941.
[11]Chaofan W,Jinsong D,Ke W,et al.Object-based classification approach for greenhouse mapping using Landsat-8imagery.International Journal of Agricultural and Biological Engineering,2016,9(1):79~88.
[12]Chen Z,Wang L,Wu W,et al.Monitoring plastic-mulched farmland by Landsat-8 OLI imagery using spectral and textural features.Remote Sensing,2016,8(4):353.
[13]Novelli A,Aguilar M A,Nemmaoui A,et al.Performance evaluation of object based greenhouse detection from Sentinel-2MSI and Landsat 8 OLI data:A case study from Almería(Spain).International Journal of Applied Earth Observation and Geoinformation,2016,52:403~411.
[14]Nemmaoui A,Aguilar M,Aguilar F,et al.Greenhouse crop identification from multi-temporal multi-sensor satellite imagery using object-based approach:A case study from Almería(Spain).Remote Sensing,2018,10(11):1751.
[15]Aguilar M,Bianconi F,Aguilar F,et al.Object-based greenhouse classification from GeoEye-1 and WorldView-2 stereo imagery.Remote sensing,2014,6(5):3554~3582.
[16]Aguilar M,Vallario A,Aguilar F,et al.Object-based greenhouse horticultural crop identification from multi-temporal satellite imagery:A case study in Almeria,Spain.Remote Sensing,2015,7(6):7378~7401.
[17]Aguilar M,Nemmaoui A,Novelli A,et al.Object-based greenhouse mapping using very high resolution satellite data and Landsat 8 time series.Remote Sensing,2016,8(6):513.
[18]元秀,晓阳.高分辨率卫星影像处理指南.北京:科学出版社,2008.
[19]Roelfsema C,Kovacs E,Ortiz J C,et al.Coral reef habitat mapping:A combination of object-based image analysis and ecological modelling.Remote sensing of environment,2018,208:27~41.
[20]Zhong Y,Wang X,Xu Y,et al.Mini-UAV-Borne Hyperspectral Remote Sensing:From Observation and Processing to Applications.IEEE Geoscience and Remote Sensing Magazine,2018,6(4):46~62.
[21]王利民,刘佳,杨玲波,等.基于无人机影像的农情遥感监测应用.农业工程学报,2013,29(18):136~145.
[22]张开生,张盟蒙.基于ZigBee的温室大棚环境监测系统研究.陕西理工学院学报(自然科学版),2015(6):18~22.
[23]韦玉春,汤国安,杨昕.遥感数字图像处理教程,2007.
[24]谷宁.基于eCognition的城市规划动态监测技术研究.北京:北京林业大学,2007.
[25]Bruzzone L,Roli F,Serpico S B.An extension of the Jeffreys-Matusita distance to multiclass cases for feature selection.IEEETransactions on Geoscience and Remote Sensing,1995,33(6):1318~1321.
