Research Of Pixel-level Image Fusion Based On Wavelet Transform

A new image fusion of multi-sensor information of multi levels, multi-level, multi-processing and comprehensive, useful information is more rich, more accurate, more reliable, more comprehensive, high quality, is conducive to the image enhancement, image segmentation, image classification, further improve the image resolution, enhance the image recognition ability, for the human decision making and artificial intelligence decision to provide basis for decision making, especially with the development of sensor technology, is widely used in automatic target recognition, intelligent robot, computer vision, remote sensing, medical image processing, manufacturing and other fields of image fusion, so the research of image fusion is a part of the image area essential.This paper uses the discrete wavelet transform, discrete wavelet frame transform and dual tree complex wavelet transform is used to transform the image, using the given fusion rule, experiment with the same goal of a series of images contain different information, and logo recognition aircraft, automobile and medicine on CT and MRI image fusion on the research, experiments show that the fusion rule can accurately extract the fused coefficients, to obtain the fusion image with better quality. In addition, according to the study of fusion rules, to evaluate the use of quality evaluation of fusion methods of performance differences on the image to obtain various kinds of fusion methods. Focusing on the influence of multi-scale decomposition, threshold and fusion rules of fusion performance.Research on image has a very broad application prospects. Pixel level image wavelet transform fusion research is a typical application of image research field based on. The research of this paper is the method of wavelet analysis in a beneficial exploration in the application of image analysis.