Image Fusion Technology And Its Application

Image fusion refers to a process, which combines complementary information captured frommultisensor so that the synthetic images with higher-quality could meet demand for specialapplication. The thesis is mainly used in around viewing system, for which image fusion isextremely important, it extracts usefull information with developed algorithms fromredundant informations in overlap regions of front, rear, left, right four images to reconstructmissing informations and could improve the clarity of panoramic image. Image preprocessingfor image fusion includes image enhancement, calibration, and the results has a direct impacton the quality of fusion image.This paper firstly discusses the background and significance of image fusion technology,makes a detailed analysis about theroy, evaluation criteria and some reference index.Secondly due to the strict requirement of algthrim efficiency in engineering applications, wemake a depth reseach about image preprocessing, then a large number of experimetns anddatas demonstrate the algorithms. In the paper we present a novel linear features basedcalibration methods for distorted image. Lastly, due to advantage of Mallat algorithm inanalyzing the local signal, we take lifting wavelet into around viewing system. Image fusionmakes the seams smooth because of different exposure conditions and weaken the ghostingbecause of the image registion error to some extent. Experiment datas about evaluation indexsof before and after image fusion demonstrate the lifting algorithm enhance the visualperception for around viewing system.