论文部分内容阅读
针对IHS变换的影像融合易产生光谱扭曲(颜色失真)的问题,文章提出了一种基于IHS变换与Bandelet变换相结合的遥感影像融合算法,利用Bandelet变换能很好捕获原始图像的边缘与纹理的特性,获取更好的彩色图像特征信息,达到最佳逼近效果:对多光谱影像Mul做IHS变换并提取I分量,通过对I分量与高分辨影像Pan做Bandelet变换,并对其几何流和Bandelet系数采用不同融合规则,再由IHS逆变换实现图像融合。实验结果表明,基于IHS与Bandelet的融合算法较IHS法在获取更丰富影像边缘细节特征信息的同时,光谱扭曲(颜色失真)现象也明显减弱,并获得了较好的人眼视觉效果,其中三角IHS与Bandelet融合的效果较好。
In order to solve the problem of spectral distortion (color distortion) in image fusion based on IHS transform, this paper proposes a remote sensing image fusion algorithm based on IHS transform and Bandelet transform. The Bandelet transform can capture the edge and texture of the original image well Characteristics and obtain better color image feature information to achieve the best approximation effect: IHS transform multi-spectral image Mul and extract I component, through the I component and high resolution Pan Pan Banlet transform, and its geometric flow and Bandelet The coefficients adopt different fusion rules, and then IHS inverse transform to realize the image fusion. The experimental results show that the fusion algorithm based on IHS and Bandelet has a significantly weaker spectral distortion (color distortion) than the IHS method while obtaining more detailed image edge feature information, and obtains better human visual effects. The triangle IHS and Bandelet fusion better.