基于PCA+àstrous小波融合方法

• 作 者: 罗艳 刘荣 朱煜峰
• 机 构: 东华理工大学地测学院,
• 出 处: 测绘科学
• 发表日期: 2009年06期
• 关键词: 影像融合 高空间分辨率 低分辨率图像 小波变换 评定准则 PCA 融合方法 融合影像 strous 融合规则

针对PCA变换融合影像存在较严重的光谱失真现象,以及àstrous小波融合影像保真度高,而空间分辨率相对低的情况,本文提出一种基于PCA+àstrous小波融合算法。新方法首先对将多波段图像经PCA变换至各不相关的成分,而后对高分辨率图像与低分辨率图像主成分按照特定融合规则进行融合处理,并使用该融合后的第一主成份分量来替代高分辨率图像与低分辨率图像进行àstrous小波融合,即PCA变换与àstrous小波变换相结合的融合处理方法。主观视觉分析和客观参数表明,新方法不仅很好的保留了影像的光谱信息,而且兼顾了地物细节能力的表达。

Aiming at the serious spectral distortion of PCA transformed fusion image and the high fidelity of àstrous wavelet fusion image, and the relatively low spatial resolution, this paper proposes a PCA + àstrous wavelet fusion algorithm. In the new method, the multi-band image is first transformed to irrelevant components by PCA, and then the main components of the high-resolution image and the low-resolution image are fused according to a specific fusion rule, and the fused first principal component Instead of high-resolution images and low-resolution images àstrous wavelet fusion, that PCA transform and àstrous wavelet transform combination of fusion processing methods. Subjective visual analysis and objective parameters show that the new method not only retains the spectral information of the image well, but also takes into account the ability to express the detail of the feature.