【摘要】 高光谱遥感图像包含了丰富的空间、辐射和光谱三重信息,它的出现是遥感界的一场革命。支持向量机在解决小样本、非线性及高维模式识别中表现出许多特有的优势,并能够推广应用到函数拟合等其他机器学习问题中。本文提出了一种新颖的水深反演方法,该方法有效地利用了高光谱遥感图像波段丰富和支持向量机在函数拟合方面的优势,将二者结合应用到海岸带浅海区水深反演的方法研究中,得到了很好的反演结果。更多还原

【Abstract】 Hyperspectral remote sensing image contains rich space,radiation and spectral information,it is a revolution in the field of remote sensing.Support vector machine shows many special advantages in solving the small sample,nonlinear and high dimensional pattern recognition,and can be applied to the function fitting other machine learning problems. This paper presents a novel water depth inversion method,this method effectively makes using of the rich band of hyperspectral remote sensing image and function fitting of support vector machine,the research combines the two to the shallow water depth inversion,the inversion results are very good.更多还原