毕业论文—基于sift算法的目标识别.doc

摘 要 目标识别是指一个特殊目标（或一种类型的目标）从其它目标（或其它类型的目标）中被区分出来的过程。 Abstract Target identification refers to the process of a particular target(a type of target) distinguish from other targets(or other types of targets). Generally speaking, target identification technology is of important, not only in the application of image processing technology, but also in the field of pattern information. Besides, it is also the basis of some other technology of image analysis. Therefore, it would seem that further investigation about target identification technology is needed. Target identification is a multistep process. Different matching algorithm would have different calculate steps and results. In addition, SIFT is an algorithm based on scale-space, which feature is the local features of the image, its rotation, scale zooming, brightness variation have the advantage of invariability, and the change of viewpoint ,pattern transformation, noise also has good trait of stability. Meanwhile, the SIFT algorithm has the characteristics of good uniqueness, huge information, faster running speed and so on. This article presents that the method of Target identification what based on SIFT algorithm, and summarize that the implementation procedure of SIFT algorithm as well as explores the matching results of SIFT in the aspect of brightness variation, scale variation and Rotation variation.Besides, it also discusses the way of object recognition by use of SIFT algorithm. Keywords: Target identification, scale-space, stability, SIFT algorithms, object recognition 目 录 第1章 绪论 1 1.1 研究背景与意义 1 1.2 国内外研究现状 2 1.3 目标识别的基本流程 4 1.4 本文主要内容 4 第2章 相关背景知识 5 2.1 数字图像处理的概念 5 2.2 数字图像处理的基本方法 6 2.3 本章小结 8 第3章 图像预处理 9 3.1 图像平滑 9 3.1.1 均值滤波 9 3.1.2 中值滤波 9 3.1.3 高斯滤波 10 3.2 图像增强 11 3.2.1 直方图均衡化 11 3.2.2 Wallis 滤波 12 第4章 SIFT特征匹配 14 4.1 图像的初始 15 4.2 尺度空间极值检测 15 4.2.1 高斯模糊 15 4.2.2 高斯金字塔 17 4.2.3 建立DoG金字塔 18 4.2.4 极值检测，初步确立特征点 19 4.3 精确定位特征点的位置和所在的尺度 20 4.3.1 关键点精确定位 20 4.3.2 去除边缘响应 21 4.4