多特征参数支持的红树林遥感信息提取——以广东省为例

doi:10.6046/zrzyyg.2022482

摘要
图/表
参考文献
相关文章
Metrics

全文: PDF(13911 KB)

HTML
输出: BibTeX | EndNote (RIS)

摘要

准确的红树林分布信息对红树林保护和管理具有重要意义。尽管已有不少红树林遥感制图研究,但如何有效利用多源遥感特征来提高红树林制图精度仍有待探索。首先,利用多源遥感数据提取光谱、散射、纹理和地形等时序特征来设计15种特征组合; 然后,利用随机森林模型分析不同特征组合在红树林识别中的精度,从而获得最优特征组合; 最后,基于Google Earth Engine(GEE)平台获取2021年广东省10 m空间分辨率的红树林分布。结果显示,冬季光谱特征的重要性最高,特征类型越丰富对应制图精度越高,最优特征组合的总体精度为92.25%,Kappa系数为0.91。通过探究红树林识别中的最优特征组合,在多特征参数支持下实现广东省红树林信息提取,研究成果可为大范围红树林精准制图提供科学参考。

	服务

	把本文推荐给朋友
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	王煜淼
	李胜
	东春宇
	杨刚

关键词 ：红树林提取, 多源遥感数据, GEE, 机器学习, 广东省

Abstract：

Accurate mangrove forest distribution information is critical to the conservation and management of mangrove forests. Despite extensive studies on the remote sensing mapping of mangrove forests, it is necessary to improve their mapping accuracy by effectively utilizing multi-source remote sensing features. First, this study designed 15 feature associations using temporal features, including spectral, scattering, texture, and terrain features, which were extracted from multi-source remote sensing data. Then, using a random forest model, it analyzed the accuracy of different feature associations in mangrove forest identification, obtaining the optimal feature association. Finally, this study mapped the 10-m-resolution mangrove forest distribution of Guangdong Province in 2021 based on platform Google Earth Engine (GEE). The results show that spectral features in winter exhibited the highest importance, with richer feature types corresponding to higher mapping accuracy. The optimal feature association yielded overall accuracy of 92.25% and a Kappa value of 0.91. Overall, this study extracted information on mangrove forests in Guangdong based on multi-feature parameters and the optimal feature association. The results of this study will provide a scientific reference for accurate mapping of mangrove forests on a large scale.

Key words： information extraction of mangrove forests multi-source remote sensing data GEE machine learning Guangdong Province

收稿日期: 2022-12-12 出版日期: 2024-03-13

ZTFLH:

TP79

基金资助:自然资源部城市国土资源监测与仿真重点实验室开放课题“联合时序SAR与光学遥感数据的广东省红树林精准识别研究”(KF-2021-06-089);宁波市重大科技攻关项目“海岸带碳库资源遥感调查与生态碳汇估算关键技术研发”(20212ZDYF020049);浙江省自然科学基金探索青年项目“基于多源遥感时序数据的大区域农作物早期识别方法研究”(LQ22D010007)

通讯作者: 杨刚(1986-),男,博士,副教授,主要从事海岸带遥感研究。Email: yanggang@nbu.edu.cn。

作者简介: 王煜淼(1992-),男,博士,助理研究员,主要从事海岸带遥感研究。Email: wymfrank@whu.edu.cn。

引用本文:

王煜淼, 李胜, 东春宇, 杨刚. 多特征参数支持的红树林遥感信息提取——以广东省为例[J]. 自然资源遥感, 2024, 36(1): 95-102.
WANG Yumiao, LI Sheng, DONG Chunyu, YANG Gang. Remote sensing information extraction for mangrove forests based on multi-feature parameters: A case study of Guangdong Province. Remote Sensing for Natural Resources, 2024, 36(1): 95-102.

链接本文:

https://www.gtzyyg.com/CN/10.6046/zrzyyg.2022482 或 https://www.gtzyyg.com/CN/Y2024/V36/I1/95

Fig.1 研究区位置及样本数据分布示意图

指数名称	计算公式^①
归一化植被指数	NDVI= $N I R - R e d N I R + R e d$
地表水指数	LSWI= $N I R - S W I R 1 N I R + S W I R 1$
修正归一化差异水体指数	MNDWI= $G r e e n - S W I R 1 G r e e n + S W I R 1$
淹没红树林指数	IMFI= $B l u e + G r e e n - 2 N I R B l u e + G r e e n + 2 N I R$
红边归一化植被指数	RENDVI= $R E 2 - R E 1 R E 2 + R E 1$

Tab.1 植被指数计算公式

Fig.2 分类精度与特征数量的关系

Fig.3 特征重要性在季节上的分布

Tab.2 组合1—4的具体特征

Fig.4 不同特征组合的识别精度

Fig.5 不同特征组合在测试集上的混淆矩阵

Tab.3 组合11—15的用户精度和制图精度

Fig.6 广东省沿海城市红树林分布与统计

Fig.7 湛江市局部红树林制图对比

[1]	Pastor-Guzman J, Dash J, Atkinson P M. Remote sensing of mangrove forest phenology and its environmental drivers[J]. Remote Sensing of Environment, 2018, 205:71-84. doi: 10.1016/j.rse.2017.11.009
[2]	何斌源, 范航清, 王瑁, 等. 中国红树林湿地物种多样性及其形成[J]. 生态学报, 2007, 27(11): 4859-4870.
	He B Y, Fan H Q, Wang M, et al. Species diversity in mangrove wetlands of China and its causation analyses[J]. Acta Ecologica Sinic, 2007, 27(11): 4859-4870.
[3]	章恒, 王世新, 周艺, 等. 多源遥感影像红树林信息提取方法比较[J]. 湿地科学, 2015, 13(2):145-152.
	Zhang H, Wang S X, Zhou Y, et al. Comparison of different metho-ds of mangrove extraction from multi-source remote sensing images[J]. Wetland Science, 2015, 13(2):145-152.
[4]	麦少芝, 徐颂军. 广东红树林资源的保护与开发[J]. 海洋开发与管理, 2005, 22(1):44-48.
	Mai S Z, Xu S J. Protection and development of mangrove resources in Guangdong Province[J]. Ocean Development and Management, 2005, 22(1):44-48.
[5]	Giri C. Observation and monitoring of mangrove forests using remote sensing:Opportunities and challenges[J]. Remote Sensing, 2016, 8(9):783. doi: 10.3390/rs8090783
[6]	Kirui K B, Kairo J G, Bosire J, et al. Mapping of mangrove forest land cover change along the Kenya coastline using Landsat imagery[J]. Ocean and Coastal Management, 2013, 83:19-24. doi: 10.1016/j.ocecoaman.2011.12.004
[7]	Warfield A D, Leon J X. Estimating mangrove forest volume using terrestrial laser scanning and UAV-derived structure-from-motion[J]. Drones, 2019, 3(2):32. doi: 10.3390/drones3020032
[8]	徐逸, 甄佳宁, 蒋侠朋, 等. 无人机遥感与XGBoost的红树林物种分类[J]. 遥感学报, 2021, 25(3):737-752.
	Xu Y, Zhen J N, Jiang X P, et al. Mangrove species classification with UAV-based remote sensing data and XGBoost[J]. National Remote Sensing Bulletin, 2021, 25(3):737-752. doi: 10.11834/jrs.20210281
[9]	王子予, 刘凯, 彭力恒, 等. 基于Google Earth Engine的1986—2018 年广东红树林年际变化遥感分析[J]. 热带地理, 2020, 40(5):881-892. doi: 10.13284/j.cnki.rddl.003268
	Wang Z Y, Liu K, Peng L H, et al. Analysis of mangrove annual changes in Guangdong Province during 1986—2018 based on Google Earth Engine[J]. Tropical Geography, 2020, 40(5):881-892.
[10]	Jia M, Wang Z, Wang C, et al. A new vegetation index to detect periodically submerged mangrove forest using single-tide Sentinel-2 imagery[J]. Remote Sensing, 2019, 11(17):2043. doi: 10.3390/rs11172043
[11]	Jia M, Wang Z, Li L, et al. Mapping China’s mangroves based on an object-oriented classification of Landsat imagery[J]. Wetlands, 2014, 34(2):277-283. doi: 10.1007/s13157-013-0449-2
[12]	Ma C, Ai B, Zhao J, et al. Change detection of mangrove forests in coastal Guangdong during the past three decades based on remote sensing data[J]. Remote Sensing, 2019, 11(8):921. doi: 10.3390/rs11080921
[13]	Xiao H, Su F Z, Fu D J, et al. 10-m global mangrove classification products of 2018—2020[DS/OL]. Science Data Bank, 2021[2022-12-09].http://doi.org/10.11922/sciencedb.01019.
[14]	邓静雯, 田义超, 张强, 等. 机载LiDAR在红树林林分平均高估算中的应用[J]. 自然资源遥感, 2022, 34(3):129-137.doi: 10.6046/zrzyyg.2021237.
	Deng J W, Tian Y C, Zhang Q, et al. Application of airborne LiDAR in the estimation of the mean height of mangrove stand[J]. Remote Sensing for Natural Resources, 2022, 34(3):129-137.doi: 10.6046/zrzyyg.2021237.
[15]	Gorelick N, Hancher M, Dixon M, et al. Google Earth Engine:Planetary-scale geospatial analysis for everyone[J]. Remote Sensing of Environment, 2017, 202:18-27. doi: 10.1016/j.rse.2017.06.031
[16]	Zhao C, Qin C Z. 10-m-resolution mangrove maps of China derived from multi-source and multi-temporal satellite observations[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2020, 169:389-405. doi: 10.1016/j.isprsjprs.2020.10.001
[17]	Peng Y, Zheng M, Zheng Z, et al. Virtual increase or latent loss? A reassessment of mangrove populations and their conservation in Guangdong,southern China[J]. Marine Pollution Bulletin, 2016, 109(2):691-699. doi: 10.1016/j.marpolbul.2016.06.083
[18]	吴培强, 马毅, 李晓敏, 等. 广东省红树林资源变化遥感监测[J]. 海洋学研究, 2011, 29(4):16-24.
	Wu P Q, Ma Y, Li X M, et al. Remote sensing monitoring of the mangrove forests resources of Guangdong Province[J]. Journal of Marine Sciences, 2011, 29(4):16-24.
[19]	江娜, 陈超, 韩海丰. 海岸带地类统计模型中DEM空间尺度优选方法[J]. 自然资源遥感, 2022, 34(1):34-42.doi: 10.6046/zrzyyg.2022005.
	Jiang N, Chen C, Han H F. An optimization method of DEM resolution for land type statistical model of coastal zones[J]. Remote Sensing for Natural Resources, 2022, 34(1):34-42.doi: 10.6046/zrzyyg.2022005.
[20]	Farr T G, Rosen P A, Caro E, et al. The shuttle Radar topography mission[J]. Reviews of Geophysics, 2007, 45(2):361.
[21]	Tucker C J. Red and photographic infrared linear combinations for monitoring vegetation[J]. Remote Sensing of Environment, 1979, 8(2):127-150. doi: 10.1016/0034-4257(79)90013-0
[22]	Xiao X, Boles S, Liu J, et al. Mapping paddy rice agriculture in southern China using multi-temporal MODIS images[J]. Remote Sensing of Environment, 2005, 95(4):480-492. doi: 10.1016/j.rse.2004.12.009
[23]	Xu H. Modification of normalised difference water index (NDWI) to enhance open water features in remotely sensed imagery[J]. International Journal of Remote Sensing, 2006, 27(14):3025-3033. doi: 10.1080/01431160600589179
[24]	贾明明. 1973—2013年中国红树林动态变化遥感分析[D]. 长春: 中国科学院东北地理与农业生态研究所, 2014.
	Jia M M. Remote sensing analysis of China’s mangrove forests dynamics during 1973 to 2013[D]. Changchun: Northeast Institute of Geography and Agroecology,Chinese Academy of Sciences, 2014.
[25]	Lin S, Li J, Liu Q, et al. Evaluating the effectiveness of using vegetation indices based on red-edge reflectance from Sentinel-2 to estimate gross primary productivity[J]. Remote Sensing, 2019, 11(11):1303. doi: 10.3390/rs11111303
[26]	何昭欣, 张淼, 吴炳方, 等. Google Earth Engine支持下的江苏省夏收作物遥感提取[J]. 地球信息科学学报, 2019, 21(5):752-766. doi: 10.12082/dqxxkx.2019.180420
	He Z X, Zhang M, Wu B F, et al. Extraction of summer crop in Jiangsu based on Google Earth Engine[J]. Journal of Geo-Information Science, 2019, 21(5):752-766.
[27]	Naboureh A, Li A, Bian J, et al. A hybrid data balancing method for classification of imbalanced training data within Google Earth Engine:Case studies from mountainous regions[J]. Remote Sensing, 2020, 12(20):3301. doi: 10.3390/rs12203301
[28]	Ghorbanian A, Kakooei M, Amani M, et al. Improved land cover map of Iran using Sentinel imagery within Google Earth Engine and a novel automatic workflow for land cover classification using migrated training samples[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2020, 167:276-288. doi: 10.1016/j.isprsjprs.2020.07.013
[29]	Song Q, Hu Q, Zhou Q B, et al. In-season crop mapping with GF-1/WFV data by combining object-based image analysis and random forest[J]. Remote Sensing, 2017, 9(11):1184. doi: 10.3390/rs9111184
[30]	Van Beijma S, Comber A, Lamb A. Random forest classification of salt marsh vegetation habitats using quad-polarimetric airborne SAR,elevation and optical RS data[J]. Remote Sensing of Environment, 2014, 149:118-129. doi: 10.1016/j.rse.2014.04.010

[1]	宋英旭, 邹昱嘉, 叶润青, 贺志霞, 王宁涛. 利用GEE云平台实现三峡库区滑坡危险性动态分析[J]. 自然资源遥感, 2024, 36(1): 154-161.
[2]	排日海·合力力, 昝梅. 干旱区绿洲城市生态环境时空格局变化及影响因子研究[J]. 自然资源遥感, 2023, 35(3): 201-211.
[3]	于森, 贾明明, 陈高, 鲁莹莹, 李毅, 张博淳, 路春燕, 李慧颖. 基于LandTrendr算法海南东寨港红树林扰动研究[J]. 自然资源遥感, 2023, 35(2): 42-49.
[4]	石敏, 李慧颖, 贾明明. 基于GEE云平台与Landsat数据的山口自然保护区红树林时空变化分析[J]. 自然资源遥感, 2023, 35(2): 61-69.
[5]	黄晓宇, 王雪梅, 卡吾恰提·白山. 基于Landsat8 OLI影像干旱区绿洲土壤含盐量反演[J]. 自然资源遥感, 2023, 35(1): 189-197.
[6]	李显风, 袁正国, 邓卫华, 杨立苑, 周雪莹, 胡丽丽. 融合多种机器学习模型的2 m气温空间降尺度方法[J]. 自然资源遥感, 2023, 35(1): 57-65.
[7]	李叶繁, 王琳, 张冬珠. 光谱特征和空间卷积相协同的近岸海域养殖塘遥感信息提取[J]. 自然资源遥感, 2022, 34(4): 42-52.
[8]	李毅, 程丽娜, 鲁莹莹, 张博淳, 于森, 贾明明. 基于最大值合成和最大类间方差法莱州湾滨海滩涂变化研究[J]. 自然资源遥感, 2022, 34(4): 68-75.
[9]	苏玮, 林阳阳, 岳文, 陈颖彪. 基于U-Net卷积神经网络的广东省海水养殖区识别及其时空变化遥感监测[J]. 自然资源遥感, 2022, 34(4): 33-41.
[10]	钞振华, 车明亮, 侯胜芳. 基于时间序列遥感数据植被物候信息提取软件发展现状[J]. 自然资源遥感, 2021, 33(4): 19-25.
[11]	王正, 贾公旭, 张清凌, 黄粤. COVID-19疫情背景下2020年第一季度广东省二、三产业GDP空间分布变化分析[J]. 自然资源遥感, 2021, 33(3): 184-193.
[12]	周询, 王跃宾, 刘素红, 于佩鑫, 王西凯. 一种遥感影像自动识别耕地类型的机器学习算法[J]. 国土资源遥感, 2018, 30(4): 68-73.
[13]	易佳思, 胡翔云. 基于Grabcut融合多源数据提取不透水面[J]. 国土资源遥感, 2018, 30(3): 174-180.
[14]	王阳明, 张景发, 刘智荣, 申旭辉. 基于多源遥感数据西藏山南地区活动断层解译[J]. 国土资源遥感, 2018, 30(3): 230-237.
[15]	侯小刚, 郑照军, 李帅, 陈雪华, 崔宇. 近15年新疆逐日无云积雪覆盖产品生成及精度验证[J]. 国土资源遥感, 2018, 30(2): 214-222.

Viewed

Full text

Abstract

Cited

Shared

Discussed