摘要图像配准是图像处理应用方面的一个基础问题,用于将不同传感器、不同视角、不同时间或不同拍摄条件下获取的同一场景的两幅或者多幅图像进行对准、叠加,在很多领域有着广泛的应用。本文是针对可见光与红外图像的配准技术的研究,常用的配准方法分为两大类:基于区域的配准方法、基于特征的配准方法。其中基于变换域的图像配准方法是一类重要的方法,如傅立叶变换域、小波变换域及Walsh变换域。本文采用的相位相关法是一种基于傅立叶功率谱的频域相关技术,先对图像进行处理,再用傅立叶变换找出偏移量从而实行有效的配准,该方法解决了仅存在平移关系的图像间的配准问题,具有很大的局限性。64878
毕业论文关键词 图像配准 傅立叶变换 变换域 相位相关法
毕业设计说明书(论文)外文摘要
Title Dual-spectral image registration techniques
Abstract
Image registration is an image processing application of a fundamental question for the different sensors,different perspectives,different times or for different shooting conditions,the same scene two or more images for the alignment,superimposed,in many areas has wide range of applications.This article is for the visible and infrared image registration technology research,common registration method is pided
into two categories:region-based registration method,feature-based registration method.Which transform domain-based image registration method is an important class of method,such as Fourier transform domain,wavelet transform domain and Walsh transform domain.Phase correlation method used in this paper is based on the Fourier power spectrum in the frequency domain correlation technique,the first image
processing,and then find the Fourier transform offsets to implement effective registration,the method to solve the translational relationship exists only registration problem between images,has serious limitation.
Keywords Image Registration Fourier Transform
Transform domain Phase correlation method
目 次
1 绪论 2
1.1 课题背景 2
1.2 图像配准的意义 2
1.3 图像配准的应用 3
2 图像配准基本理论 4
2.1 图像配准原理 4
2.2 图像配准的基本框架 4
2.3 配准方法分类 6
2.4 常用的图像配准技术 8
2.5 配准的过程 8
3 双谱图像配准的理论基础 9
3.1 可见光与红外图像特征 9
3.2 图像配准的方法分类 11
4 基于变换域的图像配准方法及MATLAB的仿真 21
4.1 傅立叶变换介绍