粒子群优化Contourlet变换的遥感影像融合算法研究

Contourlet变换能有效地应用在图像增强领域,基于Contourlet变换的遥感影像融合算法常采用高频低频替换和频域线性加权两种融合方法,因此,通过粒子群优化算法对基于Contourlet变换分解不同子带图像所需的不同加权系数进行优化,采用多元回归分析方法设定目标函数,实现了全色影像与多光谱影像的融合,与传统的Contourlet变换、PCA算法、高通滤波遥感影像融合算法相比较,新方法在提高影像清晰度的同时在光谱保真度方面相对于其他算法有明显优势.

Particle Swarm Optimization Contourlet Transform of Remote Sensing Image Fusion Algorithm Research

As a result of Contourlet transform can be effectively applied in the field of image enhancement, the characteristics of Contourlet transform in image fusion algorithm has been widely attention. Remote sensing image fusion algorithm based on Contourlet transform always utilize frequency domain linear weighted and swap high frequency to low frequency. By particle swarm optimization algorithm,the different weighting coefficients of different sub-band image by Contourlet transform is optimization. Using multivariate analysis method to set the objective function realizes the panchromatic image and multispectral image fusion. Compared with traditional Contourlet transform, (principal component analysis) algorithm, high-pass filter for remote sensing image fusion algorithm,the new algorithm with improving the definition of image have obvious advantages in terms of spectral fidelity than other methods.