基于遥感影像的沿江平原水体信息提取方法
A Method of Extracting Water Body Information along the River Plain Based on Remote Sensing Images
DOI: 10.12677/IJE.2022.112012, PDF,   
作者: 吴星麒*:浙江师范大学化学与生命科学学院,浙江 金华;张 松:中国农业大学资源与环境学院,北京;詹 超:安徽师范大学地理与旅游学院,安徽 芜湖;姚鑫鹏:安徽省六安市文峰学校,安徽 六安
关键词: TM影像遥感影像水体信息沿江平原提取TM Image Remote Sensing Image Water Body Information Plain along the River Extract
摘要: 随着航空航天技术的发展,采用卫星遥感影像提取水体信息逐渐成为研究热点。Landsat8数据产品是开放获取的中高分辨率遥感影像,选择芜湖地区进行研究,分别使用监督分类(支持向量机分类、卷积神经网络)、面向对象方法、水体指数法(NDWI、MNDWI、AWEIsh、RNDWI、EWI)提取水体分布信息,分析了不同方法的提取结果。结果表明:1) 通过Kappa检验显示出MNDWI、监督分类方法平均总体精度最高效果最好,分别为95%和94%。2) 面向对象方法分类对小区域零碎水体分类效果比较好。3) 水体在红绿波段反射率较强,而在近红外及中红外波段吸收较强,红外波段可以较好的识别地物信息。因此,从绿波段到红外波段权重递减的MNDWI水体指数法能够有效地抑制反射率噪声,在水体提取方面有较大的优势。本文对水文水资源的研究与应用具有一定的实际价值,尤其为沿江平原地区健康发展和水资源综合管理调查研究与保护提供新视野。
Abstract: With the development of aerospace technology, water information extraction from satellite remote sensing image has gradually become a research hotspot. Landsat8 data product is an open access medium and high resolution remote sensing image. Wuhu area was selected for research. Supervised classification (support vector machine classification, convolutional neural network), object-oriented method and water index method (NDWI, MNDWI, AWEIsh, RNDWI, EWI) were used to extract water distribution information. The extraction results of different methods were analyzed. The results showed as follows: 1) Kappa test showed that MNDWI and supervised classification had the highest average overall accuracy and the best effect, which were 95% and 94%, respectively. 2) The object oriented classification method has better classification effect on small area fragmentary water body. 3) The reflectivity of water body is strong in red and green bands, while the absorption is strong in near-infrared and mid-infrared bands, and the ground object information can be better recognized in the infrared band. Therefore, the MNDWI water index method with decreasing weight from green band to infrared band can effectively suppress the reflectivity noise and has great advantages in water body extraction. This paper has a certain practical value for the study and application of hydrology and water resources, especially for the healthy development of the plain area along the river and provides a new vision for the investigation and protection of integrated water resources management.
文章引用:吴星麒, 张松, 詹超, 姚鑫鹏. 基于遥感影像的沿江平原水体信息提取方法[J]. 世界生态学, 2022, 11(2): 87-95. https://doi.org/10.12677/IJE.2022.112012

参考文献

[1] 王根绪, 刘桂民, 常娟. 流域尺度生态水文研究评述[J]. 生态学报, 2005, 25(4): 892-903.
[2] Sophocleous, M. (2002) Interactions between Groundwater and Surface Water: The State of the Science. Hydrogeology Journal, 10, 52-67. [Google Scholar] [CrossRef
[3] 刘玉梅. 不同水体提取方法的提取效果比较[J]. 陕西水利, 2021(11): 104-106. [Google Scholar] [CrossRef
[4] 邰晓曼, 仲臣. Landsat-8遥感影像不同水体提取方法适用性分析[J]. 黑龙江工程学院学报, 2021, 35(4): 1-5. [Google Scholar] [CrossRef
[5] 何海清, 杜敬, 陈婷, 陈晓勇. 结合水体指数与卷积神经网络的遥感水体提取[J]. 遥感信息, 2017, 32(5): 82-86.
[6] 何予霄, 周江. ENVI中几种监督分类方法精度比较[J]. 科技创新与应用, 2018(13): 22-24.
[7] Vapnik, V.N. (1999) An Overview of Statistical Learning Theory. IEEE Transactions on Neural Networks, 10, 988-999. [Google Scholar] [CrossRef] [PubMed]
[8] 张铭飞, 高国伟, 胡静芳, 宋钰. 基于卷积神经网络的遥感图像水体提取[J]. 传感器与微系统, 2022, 41(1): 72-74, 88.
[9] 崔齐, 王杰, 汪闽, 张东, 孙艳霞. 矢量约束的面向对象高分遥感影像水体提取[J]. 遥感信息, 2018, 33(4): 115-121.
[10] 殷亚秋, 李家国, 余涛, 杨红艳, 张永红. 基于高分辨率遥感影像的面向对象水体提取方法研究[J]. 测绘通报, 2015(1): 81-85. [Google Scholar] [CrossRef
[11] McFeeters, S.K. (1996) The Use of the Normalized Dif-ference Water Index (NDWI) in the Delineation of Open Water Features. International Journal of Remote Sensing, 17, 1425-1432. [Google Scholar] [CrossRef
[12] Xu, H.Q. (2006) Modification of Normalised Difference Water Index (NDWI) to Enhance Open Water Features in Remotely Sensed Imagery. International Journal of Remote Sensing, 27, 3025-3033. [Google Scholar] [CrossRef
[13] Feyisa, G.L., et al. (2014) Automated Water Extraction Index: A New Technique for Surface Water Mapping Using Landsat Imagery. Remote Sensing of Environment, 140, 23-35. [Google Scholar] [CrossRef
[14] 杨亚莉, 赵宁, 成晓倩. 基于FLAASH模型的SPOT6卫星影像大气校正及评价[J]. 现代测绘, 2015, 38(2): 3-6+9.
[15] 闫霈, 张友静, 张元. 利用增强型水体指数(EWI)和GIS去噪音技术提取半干旱地区水系信息的研究[J]. 遥感信息, 2007, 22(6): 62-67.