基于GEE平台形态学变换的枣树种植面积提取——以麦盖提县为例  

Extraction Extraction of jujube planting area based on morphological transformation of GEE platform--A case study of Maigaiti County

作　　者：王永东 尼格拉·吐尔逊 郑江华[1,3] 王蕾 罗磊 WANG Yongdong;TUERXUN Nigela;ZHENG Jianghua;WANG Lei;LUO Lei(College of Geography and Remote sensing Sciences,Xinjiang University,Urumqi 830046,China;Institute of Modern Forestry,Xinjiang Academy of Forestry Sciences,Urumqi 830000,China;Xinjiang Key Laboratory of Oasis Ecology,Xinjiang University,Urumqi 830046,China)

机构地区：[1]新疆大学地理与遥感科学学院,新疆乌鲁木齐830046 [2]新疆林业科学院现代林业研究所,新疆乌鲁木齐830000 [3]新疆绿洲生态重点实验室,新疆乌鲁木齐830046

出　　处：《甘肃农业大学学报》2025年第1期213-223,共11页Journal of Gansu Agricultural University

基　　金：新疆特色林果业典型树种——红枣树种植面积与主要健康参数遥感反演(2020670007);新疆特色林果业典型树种——红枣分布遥感监测与种植面积估算(202105140018)。

摘　　要：【目的】准确的枣树种植面积提取与空间分布制图对于枣树种植区的管理极为重要,麦盖提县是新疆主要的枣树种植区,通过遥感技术快速提取枣树种植面积对麦盖提县的农业发展和相关辅助决策具有重要作用。同时在遥感云计算平台探索遥感卫星影像与“膨胀”“腐蚀”的形态学变换相结合的作物面积提取方法,在遥感技术应用方面具有研究意义。【方法】本研究使用Google Earth Engine遥感云计算分析平台,探索了基于枣树生育期的Sentinel-2影像与形态学变换相结合在枣树种植面积提取中的应用潜力。【结果】麦盖提县枣树种植面积达37800 hm^(2),主要分布在麦盖提镇周边区域,验证提取的枣树种植面积,Kappa系数达0.82。用户精度、生产者精度和总体精度分别达82.4%、86.2%、84.3%。Sentinel-2遥感数据可以作为县域尺度枣树种植面积监测的重要数据源,“腐蚀”与“膨胀”相结合的形态学变换使提取的枣树分布与面积信息更接近实际。利用Google Earth Engine云计算可以高效地提取枣树面积信息,提高了信息提取效率,省去了影像下载、预处理等繁琐的本地处理工作。【结论】本研究基于Sentinel-2影像提取研究区枣树面积37800 hm^(2),麦盖提县枣树种植在绿洲中部分布较为集中,主要分布在麦盖提镇以北、以南区域,而在绿洲北部以西、以东区域也有枣树种植,相对稀疏,其他区域枣树种植比较零散。基于GEE云平台对提取结果进行形态学变换能提升研究区枣树种植面积的提取精度。同时本研究探索了Sentinel-2影像在枣树作物的面积提取中的应用潜力,Sentinel-2遥感数据可以作为县域尺度枣树种植面积监测的数据源。【Objective】Accurate jujube planting area extraction and spatial distribution mapping are ex⁃tremely important for the management of jujube planting area,and the rapid extraction of jujube planting area through remote sensing technology plays an important role in the agricultural development and related auxiliary decision-making of Maigaiti County.The study aimed to develop the method of crop area extrac⁃tion by combining remote sensing satellite images with morphological transformation of"expansion"and"corrosion"on the remote sensing cloud computing platform,which is of significance in the application re⁃search of remote sensing technology.【Method】The Google Earth Engine remote sensing cloud computing analysis platform was used to explore the application potential of combining Sentinel-2 images in the growth period of jujube trees with morphological transformation in the extraction of jujube planting area.【Result】The planting area of jujube trees in Maigaiti County reached 37800 hm^(2),mainly distributed in the surrounding area of Maigaiti Town,the seat of the county-level government of Maigaiti County.For the ex⁃tracted planting area of jujube trees,the Kappa coefficient was 0.82.The user accuracy,producer accuracy and overall accuracy reached 82.4%,86.2%and 84.3%,respectively.Sentinel-2 remote sensing data could be used as an important data source for monitoring the planting area of jujube trees at the county scale,and the morphological transformation combining"corrosion"and"expansion"made the extracted information on jujube distribution and area closer to the reality.The use of Google Earth Engine cloud computing could efficiently extract jujube tree area information,improve the efficiency of information extraction,and elimi⁃nate cumbersome local processing work such as image download and preprocessing.【Conclusion】Based on Sentinel-2 image extraction,the study area covered 37800 hm^(2) of jujube trees.Jujube trees in Maigaiti County were more concentrated in the middle of oasis,mainly

关 键 词：枣树 Google Earth Engine 种植面积 提取 形态学变换 

分 类 号：P962[天文地球—自然地理学]