Research on water extraction method based on GF-5 hyperspectral feature analysis
-
摘要: 针对山区水体、山体阴影与裸地等地类光谱混淆性，基于高分五号(GF-5)影像数据，结合高光谱特征分析构建了山区水体决策树提取模型. 先对水体和相关干扰地类进行高光谱特征分析实现特征波段选取，应用单波段阈值法、多波段谱间关系法、归一化水指数(NDWI)法进行提取实验. 通过比较以上实验不足之处，提出了单波段阈值法与构建的阴影水体指数(SWI)相结合的决策树水体提取模型，以Google Earth高清影像为参考结合实地采样得到的混淆矩阵进行精度评价. 实验结果表明：单波段阈值法与NDWI法易将山体阴影识别为水体，受裸地影响较小；多波段谱间关系法对山体阴影有一定抑制作用，受小面积裸地影响；决策树提取模型能有效抑制山体阴影和裸地影响提取完整水体. 其总体精度为89.39%，Kappa系数为0.82，显著提升了山区水体提取精度.
-
关键词:
- 高分五号(GF-5) /
- 高光谱特征分析 /
- 阴影水体指数 /
- 决策树
Abstract: In view of the spectral confusion between mountain water body, mountain shadow and bare land, a decision tree extraction model for mountain water body is constructed based on Gao Fen-5(GF-5) image data combined with hyperspectral feature analysis. First, perform hyperspectral feature analysis on the water body and related interference types to realize the selection of feature bands, apply single-band threshold method, multiple-band spectral relationship method, and NDWI method to extract experiments. By comparing the deficiencies of the above experiments, a single-band threshold is proposed. The decision tree water body extraction model combined with the constructed shadow water index (SWI) is used to evaluate the accuracy of the confusion matrix obtained by using Google Earth high-definition images as a reference and on-site sampling. The experimental results show that the single-band threshold method and NDWI method can easily identify mountain shadow as water body and are less affected by bare land; the multiple-band spectrum relationship method has a certain inhibitory effect on mountain shadow and is affected by small areas of bare land; decision tree model can effectively suppress the influence of mountain shadow and bare land to extract complete water body. The overall accuracy is 89.39%, and the Kappa coefficient is 0.82, which significantly improves the extraction accuracy of mountain water body. -
表 1 AHSI系统技术规格
有效波
段个数光谱范
围/nm空间分
辨率/m太阳方
位角/（º)太阳高
度角/(º)幅宽/km 光谱分
辨率/nm305 400～
250030 214.083 27.648 60 VNIR: 5 SWIR: 10 
下载: 导出CSV
表 2 不同提取方法精度验证
方法 总体精度/% Kappa系数 单波段阈值法 85.67 0.78 多波段谱间关系阈值法 78.83 0.68 NDWI 83.19 0.75 决策树法 89.39 0.82
下载: 导出CSV
-
[1] 何海清, 杜敬, 陈婷, 等. 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. DOI: 10.3969/j.issn.1000-3177.2017.05.013 [2] 朱长明, 骆剑承, 沈占锋, 等. DEM辅助下的河道细小线性水体自适应迭代提取[J]. 测绘学报, 2013, 42(2): 277-283. [3] 李丹, 吴保生, 陈博伟, 等. 基于卫星遥感的水体信息提 取研究进展与展望[J]. 清华大学学报(自然科学版), 2020, 60(2): 147-161. [4] 邱煌奥, 程朋根, 甘田红, 等. 多光谱遥感影像湿地水体提取方法综述[J]. 江西科学, 2016, 34(1): 60-65, 144. [5] 王瑗, 盛连喜, 李科, 等. 中国水资源现状分析与可持续发展对策研究[J]. 水资源与水工程学报, 2008, 19(3): 10-14. [6] WAN L, LIU M, WANG F, et al. Automatic extraction of flood inundation areas from SAR images: a case study of Jilin, China during the 2017 flood disaster[J]. International journal of remote sensing, 2019, 40(9): 1-28. DOI: 10.1080/01431161.2019.1577999 [7] 白翠, 向洋, 邱春霞, 等. 基于多源遥感数据的水体提取方法研究[J/OL]. (2021-01-27)[2021-02-18]. 人民黄河, 2021, 43(1). https://kns.cnki.net/kcms/detail/41.1128.TV.20210127.0951.002.html [8] 姜腾龙, 赵书河, 肖鹏峰, 等. 基于光谱夹角的水体信息提取方法研究[J]. 国土资源遥感, 2009, 80(2): 102-105. [9] 王志辉, 易善桢. 不同指数模型法在水体遥感提取中的 比较研究[J]. 科学技术与工程, 2007, 7(4): 534-537. DOI: 10.3969/j.issn.1671-1815.2007.04.028 [10] LI Q, MAO F, HUANG J X, et al. Extracting wetland information from SPOT5 imagery in Nansihu Area of Shadong province[C]//2008 IEEE International Geoscience and Remote Sensing Symposium, 2008. DOI: 10.1109/IGARSS.2008.4779861 [11] GENELETTI D, GORTE B G H. A method for object-oriented land cover classification combinng Landsat TM data and aerial photographs[J]. International journal of remote sensing, 2003, 24(6): 1273-1286. DOI: 10.1080/01431160210144499 [12] 邓劲松, 王珂, 李君, 等. 决策树方法从SPOT-5卫星影像中自动提取水体信息研究[J]. 浙江大学学报(农业与生命科学版), 2005, 31(2): 171-174. [13] 王培培. 基于ETM影像的水体信息自动提取与分类研究[J]. 首都师范大学学报(自然科学版), 2009, 30(6): 75-79. [14] 钟春棋, 曾从盛, 柳铮铮. 基于谱间特征与比值型指数的水体影像识别分析[J]. 地球信息科学, 2008, 10(5): 663-669. [15] 陈静波, 刘顺喜, 汪承义, 等. 基于知识决策树的城市水体提取方法研究[J]. 遥感信息, 2013, 28(1): 29-33, 37. DOI: 10.3969/j.issn.1000-3177.2013.01.007 [16] WANG J Z, DING J L, LI G N, et al. Dynamic detection of water surface area of ebinur lake using multi-source satellite data (landsat and sentinel-1A) and its responses to changing environment[J]. Catena, 2019(177): 189-201. DOI: 10.1016/j.catena.2019.02.020 [17] 周成虎, 杜云艳, 骆剑承. 基于知识的AVHRR影像的水体自动识别方法与模型研究[J]. 自然灾害学报, 1996, 6(3): 100-108. [18] JAWAK S D, LIUS A J. A rapid extraction of water body features from antarctic coastal oasis using very high-resolution satellite remote sensing data[J]. Aquatic procedia, 2015(4): 125-132. DOI: 10.1016/j.aqpro.2015.02.018 [19] 杨树文, 薛重生, 刘涛, 等. 一种利用TM影像自动提取细小水体的方法[J]. 测绘学报, 2010, 39(6): 611-617. [20] 杨存建, 徐美, 黄朝永, 等. 遥感信息机理的水体提取方法探讨[J]. 地理研究, 1998, 17(增刊): 86-89. [21] MCFEETERS S K. The use of normalized difference water index (NDWI) in the delineation of open water features[J]. International journal of remote sensing, 1996, 17(7): 1425-1432. DOI: 10.1080/01431169608948714 -
点击查看大图
图(5) / 表(2)
计量
- 文章访问数: 706
- HTML全文浏览量: 221
- PDF下载量: 63
- 被引次数: 0
