高分辨率SST反演中的图像预处理和算法修订  

Image pre-processing and algorithm revision during high-resolution SST retrieval

作　　者：黄路 郭晓明[1] 张彩云[2] 吴易达[3] 乔正明 HUANG Lu;GUO Xiaoming;ZHANG Caiyun;WU Yida;QIAO Zhengming(Preparatory Office of Zhangzhou Base,National Ocean Technology Center,Xiamen 361007,China;College of Ocean and Earth Sciences,Xiamen University,Xiamen 361102,China;91937 Troops,Zhoushan 316002,China)

机构地区：[1]国家海洋技术中心漳州基地筹建办公室,福建厦门361007 [2]厦门大学海洋与地球学院,福建厦门361102 [3]91937部队,浙江舟山316002

出　　处：《海洋测绘》2022年第6期70-73,78,共5页Hydrographic Surveying and Charting

基　　金：国家重点研发计划(2016YFE0202100);国家自然科学基金重点项目(U1805241)。

摘　　要：水陆分离、云检测及算法适应性修订等前期处理过程对于提高高分辨率海表温度(sea surface temperature,SST)遥感反演精度至关重要,为解决这部分内容长期被忽视且缺乏系统性总结等问题,首先梳理了热红外图像预处理的一般流程,重点介绍了开展水陆分离和云检测的时机、原理、方法以及应着重关注的对象等;其次给出了主流SST反演算法应用在新传感器上时须进行修订的参数及方法等,并对下步各算法改进重点方向提出建议。本研究可为高分辨率SST业务化反演提供参考。To address the problem that pre-processing such as water-land separation,cloud detection,and algorithm adaptation revision are often ignored and lack of systematic discussion for retrieving the SST with high spatial resolution.this paper analyses the pre-processing processes of thermal infrared images such as water-land separation and cloud detection,and summarizes the parameters to be revised during algorithm migration and their revision methods.It is pointed out that to improve the accuracy of SST retrieval,firstly,strict and detailed water-land separation and cloud detection should be carried out both before and after SST retrieval,focusing on the interference of non-water body pixels such as small-scale islands,ships,and thin clouds,etc.Secondly,when the algorithms are applied to a new sensor,some parameters must be revised,and the selection of a suitable atmospheric profile and interval fitting are crucial.Finally,improving the accuracy of reanalysis data sets and solving the problem of a general reduction of algorithm accuracy under high atmospheric water vapor content should be the focus of the next work.This study can provide a reference for the improvement and operational application of high-resolution SST retrieval algorithms.

关 键 词：海洋遥感 水陆分离 云检测 算法适应性修订 海表温度 热红外图像 

分 类 号：P236.2[天文地球—摄影测量与遥感]