摘要近些年来，伴随着科技发展的日新月异，计算机处理技术得到了飞速提高，图像配准算法不断涌现。本文主要讨论的是基于局部图像特征的配准算法，以实现图像快速、高效、准确地配准。

本文主要工作：63871

1.对基于局部特征的检测器和特征描述符的原理、性能进行了深入分析和对比，并总结其稳定性、结合性等方面的性能表现。为后续设计配准算法，选择适当的检测器和描述符提供了依据；

2.研究了一种快速有效的SURF算法。该算法在积分图像的基础上进行快速运算,并利用箱式滤波器建立尺度空间,通过快速Hessian矩阵检测极值点,利用哈尔小波生成描述子,采用快速索引进行相似性度量。研究了另外一种BRIEF算法，该算法将二进制串作为该特征点的特征描述子，运用汉明距离，构成和匹配非常快，是图像配准的一大进步；

3.基于上述两种算法开展了相关实验。由于BRIEF算法不具有旋转不变性、尺度不变性，对噪声不敏感，实验将BRIEF算法和SURF算法做结合，采用了SURF算法的取点方式，这样就使得BRIEF性能更加良好。另外，将所研究的算法与SIFT进行了对比实验和分析。

毕业论文关键词  图像配准、局部特征、SIFT、SURF、BRIEF等

毕业设计说明书（论文）外文摘要

Title    The image local binary feature descriptor research   and application                                       

Abstract In recent years, along with the development of science and technology, computer processing technology has been rapidly improved.Image matching algorithm continue to emerge. This article focuses on the matching algorithm based on local image features to achieve a fast, efficient and accurate match. The main work:

    First,most kinds of detectors and descriptors are described and summarized in a comprehensive way, and draw its performance,stability and binding properties and so on,to provide a theoretical basis to select the appropriate detector and descriptor for the registration algorithm.

    Second,I studied a fast and efficient SURF algorithm.SURF is computed on the integral image,uses several approximations,such as box filter in the process of establishment of scale space,fast Hessian in the process of detect key points,Haar wavelet when the descriptors are generated,and fast index in the process of similarity measures.Also I studied  BRIEF algorithm.The algorithm uses binary strings as an efficient feature point descriptor and uses the Hamming distance.the matching speed is very fast.BRIEF is a big step in image registration.

    Third,I carried out related experiments based on the two algorithms.As BRIEF algorithm does not have rotational invariance,scale invariance,not sensitive to noise,we will do BRIEF and SURF algorithm combined.SURF algorithm is used to take a point,so that makes BRIEF more favorable.

Keywords  image registration、local features、SIFT、SURF、BRIEF 

1  绪论 1

1.1  选题意义 1

1.2  图像配准概述 1

1.3  图像配准原理 1

1.4  图像配准方法 2

1.5  国内外图像配准技术的研究现状和研究意义 5

1.6  本论文主要内容 5

2  图像局部特征 7

2.1  概述 7

2.2  特征点检测