Show simple item record

dc.contributor.author吴曲然
dc.contributor.author胡建宇
dc.contributor.author孙振宇
dc.contributor.author朱佳
dc.date.accessioned2016-05-17T02:29:09Z
dc.date.available2016-05-17T02:29:09Z
dc.date.issued2015-3-28
dc.identifier.citation厦门大学学报(自然科学版),2015,(2):57-64
dc.identifier.issn0438-0479
dc.identifier.otherXDZK201502009
dc.identifier.urihttps://dspace.xmu.edu.cn/handle/2288/104816
dc.description.abstract统计模型检测法为检测海表温度(SST)锋面和提取SST锋面的结构提供了有利的工具,但其在陆架区的检测成功率和检测精度仍有待提高.本研究对该算法进行如下改进:1)在数据预处理阶段,通过梯度场拟合计算不同水团之间过渡区的边界,并利用边界处的温度数据重构过渡区两侧的温度场;2)利用遗传算法为锋面参数的迭代求解提供初值.理想实验结果表明:上述改进能有效地改善统计模型检测法在陆架区的稳定性和检测精度.最后,利用实测SST数据和遥感SST数据检验了算法在实际应用中的改进效果,以及改进后算法在陆架区的有效性.
dc.description.abstractThe statistical modeling approach to ocean front detection provides an useful tool for detecting sea surface temperature(SST)fronts and extracting the structure of SST fronts.But the robustness and accuracy of the algorithm still need to be improved when it is applied to the continental shelf.In this paper,the authors propose two improvements to the algorithm.1)On the stage of data preparation,fitting of gradient field is used to calculate boundaries of the transition zone between different water masses.Then,the temperature fields on both sides of the transition zone are reconstructed through the data extracted from the boundaries.2)Genetic algorithm is used to provide initial values for iterations.The idealized experiments show that the improved algorithm is robust to highly dynamic fronts and more accurate at the continental shelf.At last,in situ and remote sensing SST data are used to verify the improvements of the algorithm in practical application and the validity of the fronts detected by the improved algorithm at the continental shelf.
dc.description.sponsorship国家重点基础研究发展计划(973)项目(2009CB421208); 国家自然科学基金(41276006;41121091)
dc.language.isozh_CN
dc.subject海表温度峰
dc.subject卫星遥感
dc.subject自动检测
dc.subject南海北部
dc.subjectsea surface temperature front
dc.subjectsatellite remote sensing
dc.subjectautomatic detection
dc.subjectthe northern South China Sea
dc.title海洋锋面统计模型检测法的改进与验证
dc.title.alternativeImprovement and Validation of a Statistical Modeling Approach to Ocean Front Detection
dc.typeArticle


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record