Image Fusion (I.)--Simple weighted fusion

　　

Simple weighted fusion, also called Pixel weighted averaging (Weighted Averaging,wa), is the simplest and most straightforward method of image fusion. It has the advantages of simple implementation, fast operation speed, and can improve the signal-to-noise ratio of fused image, but this method weakens the detail information in the image, reduces the contrast of the image, and makes the edges in the image blurred in a certain degree, so it is difficult to obtain the satisfied fusion effect in most applications.

Optimization: Principal component Analysis (Principal Component ANALYSIS,PCA) is a commonly used coefficient optimization method, and a fused image with maximum luminance variance can be obtained by using the weights determined by principal component analysis. The PCA method is applied to the fusion of high-resolution panchromatic image and low-resolution multispectral image, and the first principal component extracted from low-resolution multispectral image is replaced by high-resolution panchromatic image, and high spatial resolution and high spectral score are obtained simultaneously.

In terms of performance, the principal component analysis method is more like the selection of the source image rather than the integration of the significant information in the source image. Limitations: Using global variance as a measure of information significance usually assigns larger weights to source images with larger variances. In practical application, when the output image of a certain sensor is low in contrast, this method of weight distribution is more effective, but in general, this method is not scientific. In addition, principal component Analysis (PCA) is very sensitive to the interference information such as dead center and noise in the image, which can significantly improve the global variance of the image.

Image Fusion (I.)--Simple weighted fusion