论文部分内容阅读
绿潮作为一种新型的海洋灾害,已经引起了各个国家的重视。依据2012年南黄海海域浒苔遥感监测分布面积数据,选取了温度、天气状况、风向、风力、浪高5种影响浒苔扩散的气候因子,建立了基于SVR的浒苔分布面积预测模型,并与经典的最近邻点插值模型、线性插值模型、3次样条函数插值模型和分段3次Hermite插值模型进行了回归效果的对比。分析结果表明,基于SVR的浒苔分布面积预测模型能够为浒苔遥感数据的插补提供一种方法,且回归效果优于传统的回归方法,为浒苔的防治提供辅助决策信息。
As a new type of marine disaster, the green tide has drawn the attention of all countries. According to the distribution data of remote sensing monitoring by Enteromorpha in the South Yellow Sea in 2012, five climatic factors influencing the spread of Enteromorpha were selected, such as temperature, weather condition, wind direction, wind force and wave height, and the prediction model of the distribution area of Enteromorpha prolifera Compared with the classical nearest neighbor interpolation model, the linear interpolation model, the cubic spline function interpolation model and the segmented cubic Hermite interpolation model, the regression results were compared. The results showed that the prediction model of Enteromorpha prolifera based on SVR can provide a method for the interpolation of the Enteromorpha remote sensing data, and the regression effect is better than the traditional regression method, providing decision support information for Enteromorpha prolifera.