基于金字塔变换的图像融合方法

图像融合的目的是综合各个图像的互补信息,合并成一个新图像来改善图像的视觉效果,获得对同一场景更为准确、全面、可靠的图像描述.目前图像融合中常用的方法有加权平均法、调制法、小波变换法、HIS假彩色法等.文中提出一种新的图像融合方法:首先利用拉普拉斯金字塔分解方法对原始图像进行多分辨率分解,然后采用基于区域特征量测的方法对分解后的各层图像进行融合,在此基础上,对融合金字塔做拉普拉斯金字塔反变换得到最后的融合图像.通过实验,证明了该方法能够取得很好的融合效果,融合质量优于其他几种方法.

An Image Fuse Method Based on Pyramid Transformation

The aim of the image fusion is to integrate the complementary information of each image.By uniting the useful information a new image can be got to improve the visual impression.Now,the main methods in image fusion are proportional average method,modulate method,wavelet transform method,HIS feigned multicolor method,and so on.This paper adopts a Laplacian pyramidal decomposition method to analyze two images,and fuse the decomposed images with the method based on regional features measurement.By doing the reversed Laplacian pyramidal transform in the next step,the final image is got.The experiment proves that this method can get a standout fusion result.The quality of the fused image is better than the results of other methods.