基于传感器光谱特性与分类的遥感图像融合

Fusion of Remote Sensing Images Based on Sensor Spectral Response and Classification

摘要:

为了提高多光谱图像和全色图像的融合质量,提出了一种基于广义亮度色度饱合度（IHS）变换与分类思想相结合的遥感图像融合算法.在充分考虑各波段传感器的相对光谱响应的基础上,在强度分量的构造和空间细节信息的注入方面进行了改进.改进后的方法首先对图像进行分类,然后采用高斯概率密度函数对分类构造后的影像进行加权融合,有效地改善了光谱畸变问题,且能同时对所有的光谱波段进行融合.以资源卫星IKONOS影像进行实验,结果表明该方法不仅能使融合图像具有较好的空间分辨率,还能很好地保持原有多光谱图像的光谱响应特性,具有潜在的应用前景.

Abstract:

In order to improve the fusion quality of the multispectral images and panchromatic images,a new fusion algorithm based on the generalized IHS transformation and supervised classification is proposed.Taking into account the spectral response of the multipectral and panchromatic sensors,the generalized HIS fusion can be improved in two parts:the construction of intensity component and injection method of detail information.In the proposed method,the images are classified into different parts,and then the Gaussian probability density function is used to do the weighted fusions to the images after the classification.These works reduce significantly the spectral distortion and all the multispectral bands can be fused at the same time.IKONOS images is taken as an example to evaluate the performance and efficiency of the proposed method.The result shows that the proposed algorithm can provide high spatial resolution images,as well as the spectral features based on original multispectral images.This algorithm proves to be prospective in the application for image fusion.