白冠统计物理模型的验证——Ⅰ.二维等间距投影图的数字化方法研制与应用

发布时间：2018-04-21 09:56

  本文选题：图像处理 + SST图像数字化和检验　； 参考：《海洋科学进展》2017年03期

【摘要】：近几十年以来,随着海洋观测技术的提高,海洋观测数据大量增加,极大地推动了海洋研究的发展,但由于某些原因,部分数据只以图像形式对外发布,很难获取其原始数据,这给科研工作的及时开展带来了诸多不便。本文基于MATLAB软件的图像和数据处理功能,提出了一种二维等间距投影图像的数字化方法,可实现图像信息快速准确的抓取,并以Levitus的全球海表温度为例,验证了数字化过程的可靠性。数字化结果与原始数据比较显示,平均误差为0.147℃,绝对平均误差为0.215℃,均方根误差为0.275℃,数字化结果与原始数据吻合较好。最后,本文将该方法应用到2001-02,2001-05,2001-08和2001-11卫星反演的白冠覆盖率图像中,所获数字化结果将应用到下一步统计物理白冠模型的验证工作中。
[Abstract]:In recent decades, with the improvement of ocean observation technology, ocean observation data have been greatly increased, which has greatly promoted the development of ocean research. However, for some reasons, some of the data are only released in the form of images, so it is difficult to obtain the original data. This brings a lot of inconvenience to the timely development of scientific research work. Based on the image and data processing function of MATLAB software, this paper presents a digitization method of 2-D equidistant projection image, which can quickly and accurately capture the image information, and take the global sea surface temperature of Levitus as an example. The reliability of the digital process is verified. Compared with the original data, the digital results show that the average error is 0.147 鈩，

本文编号：1781933