| 摘要: |
| 南海珊瑚岛礁总体处于退化状态,卫星遥感可以实现珊瑚礁区底质演变的监测。本文以西沙群岛海域赵述岛为研究区域,以Landsat-8卫星遥感影像为研究数据,对浅海地貌变化进行遥感分析。将影像进行水深校正和波段组合,使用最大似然法进行珊瑚礁地貌分类,可以清晰的识别灰沙岛、礁坪、礁前坡、泻湖坡、暗滩和海面六种地貌类型,以WorldView-2卫星影像为验证数据,总体分类精度为94.81%,高于直接使用最大似然法分类的92.30%,再结合决策树分类对分类结果中灰沙岛地貌进行改正,总体分类精度提升到95.07%;基于2014-2018年的Landsat-8数据,通过转移矩阵完成珊瑚礁地貌类型的时空变化分析,结果表明赵述岛珊瑚礁退化和转移明显。 |
| 关键词: 珊瑚礁 最大似然法 决策树分类法 水深校正 时空变化 |
| DOI：10.11759/hykx20210727004 |
| 分类号:TP79 |
| 基金项目:国家重点研发计划海洋光学遥感探测机理与模型研究(2016YFC1400904) |
|
| Remote sensing recognition of coral islands and reefs in the South China Sea |
|
YUE Zi-lin1, ZHU Wei-dong1,2, QIU Zhen-ge1,2, LUAN Kui-feng1,2, YE Li1, QIAN Chu-yi1
|
|
1.College of Marine Sciences, Shanghai Ocean University, Shanghai 201306, China;2.Shanghai Engineering Research Center of Estuarine and Oceanographic Mapping, Shanghai 201306, China
|
| Abstract: |
| Coral reefs in the South China Sea are generally in a state of degradation,and satellite remote sensing can monitor the evolution of coral reef sediments.This paper considers the Zhaoshu Island in the Paracel Islands area as the research area and takes a Landsat-8 satellite remote sensing image as the research data.The image is used to correct the depth of the water and combine the band.The reef landforms are classified using the maximum likelihood method.Six geomorphological types,namely gray sand island,reef flat,reef front slope,lagoon slope,beach,and sea surface,can be clearly identified.Taking the Worldview-2 satellite image as the verification data,the overall classification accuracy is 94.81%,which is higher than 92.30% of the maximum likelihood method.Combined with the decision tree classification,the overall classification accuracy is improved to 95.07%.The temporal and spatial changes of coral reef geomorphic types were analyzed by the transfer matrix,based on the Landsat-8 data from 2014 to 2018.Results show that the coral reef degradation and transfer are obvious in Zhaoshu Island. |
| Key words: coral reef maximum likelihood method decision tree classification water depth correction spatiotemporal change |
