中国科学院沈阳自动化研究所机构知识库
Advanced  
SIA OpenIR  > 光电信息技术研究室  > 学位论文
题名: 基于直线段编组的图像匹配方法研究
其他题名: Image matching based on line segment grouping
作者: 王学娟
导师: 罗海波
分类号: TP391.41
关键词: 边缘检测 ; 直线段提取 ; 直线段编组 ; 图像匹配
索取号: TP391.41/W37/2012
学位专业: 控制理论与控制工程
学位类别: 硕士
答辩日期: 2012-05-28
授予单位: 中国科学院沈阳自动化研究所
学位授予地点: 中国科学院沈阳自动化研究所
作者部门: 光电信息技术研究室
中文摘要: 图像匹配作为一种图像分析和处理技术越来越广泛地应用于计算机视觉、自动目标跟踪、自动目标识别、立体视觉、图像检索等领域,因此无论是在军用上还是民用上,数字图像匹配技术都是一项很重要的技术。目前基于灰度值及变换域的图像匹配技术计算量大,而基于特征的图像匹配缺少更有效的描述方法,针对上述问题,同时考虑到人造目标大多可以用直线段表征,本文提出了一种更为智能、更具有语义信息的直线段编组特征,以实现对目标或图像的科学描述。 本文首先从图像的梯度方向直方图出发提出一种采用HOG特征的直线段提取方法,该方法利用梯度直方图确定可能存在的直线方向;利用矩阵的行、列积分运算实现对原始图像特定方向直线段的投影;根据矩形函数的一阶导数性质将确定直线段端点位置的问题转化为对列方向向量的求导问题,以求能更准确地提取直线段。通过本文方法与基于区域生长及基于Hough变换的直线段提取方法的仿真实验对比表明,本文提出的直线段提取方法可有效解决双边缘及直线段断裂等问题。 在此基础上,提出了一种直线编组特征——基于仿射不变的二维特征描述矩阵,详细介绍了其构造方法,从成像的角度对其合理性进行了说明,并探讨基于射影不变性的二维特征描述矩阵构造方法。研究了不同的相似性度量方法,基于本文提出的的直线段编组特征,选用最适合的方法对基准图与实时图进行匹配。理论和实验结果表明,本文提出的直线编组特征具有光照不变性,几何不变性,可以实现更具语义信息的目标描述,并且可以实现图像间的匹配。 本文的研究成果丰富和扩展了现有的图像匹配方法,为基于特征的图像理解提供了一定的科学依据。
英文摘要: As a kind of image analysis and processing technology, Image matching is more and more widely applied to computer vision, automatic target tracking, automatic target recognition, stereo vision, image retrieval and other fields, and digital image matching technology have become a very important technology either in military, or civil. This paper proposes a straight segment grouping feature in order to realize a more semantic information description of the target or the image, aiming at the fault of large calculation of image matching method base on the grey value and transform domain and the lack of more effective significance image matching method based on the characteristics. Firstly, a method is needed to extract line segments accurately for line segments grouping, so this paper represented a HOG-feature-based approach. In this method, in order to extract line segments efficiently, the integration on rows and columns of a matrix is used to realize the projection of line segments with particular direction in original image and then with the application of the property of the rectangular function’s first derivative, the line-segment endpoints positions are determined via the column vector’s derivative. The comparison of the three methods simulation experiments have verified that the HOG-feature-based method can not only extract straight segments accurately, but also solve the straight segment fracture problems existing in the classic methods to some extent Secondly, line segments grouping feature is presented. We introduce how to constitute 2-d characteristics description matrix based on affine invariant in details, and explore description matrix with 2-d projective invariance. Then its rationality is explained from the point of view of imaging. At last, we match the reference image and the real-time image with a the most suitable method for the group-line feature after we study different similarity measure. The theoretical and experimental results show that the group-line feature is light invariant and geometry invariant, and it is a description with semantic information and can be used in image matching. The result of this paper riches and expands the existing image matching methods, and provides some scientific evidence for image understanding based on the characteristics.
语种: 中文
产权排序: 1
内容类型: 学位论文
URI标识: http://ir.sia.cn/handle/173321/9242
Appears in Collections:光电信息技术研究室_学位论文

Files in This Item:
File Name/ File Size Content Type Version Access License
基于直线段编组的图像匹配方法研究.pdf(3977KB)----限制开放 联系获取全文

Recommended Citation:
王学娟.基于直线段编组的图像匹配方法研究.[硕士 学位论文 ].中国科学院沈阳自动化研究所 .2012
Service
Recommend this item
Sava as my favorate item
Show this item's statistics
Export Endnote File
Google Scholar
Similar articles in Google Scholar
[王学娟]'s Articles
CSDL cross search
Similar articles in CSDL Cross Search
[王学娟]‘s Articles
Related Copyright Policies
Null
Social Bookmarking
Add to CiteULike Add to Connotea Add to Del.icio.us Add to Digg Add to Reddit
所有评论 (0)
暂无评论
 
评注功能仅针对注册用户开放,请您登录
您对该条目有什么异议,请填写以下表单,管理员会尽快联系您。
内 容:
Email:  *
单位:
验证码:   刷新
您在IR的使用过程中有什么好的想法或者建议可以反馈给我们。
标 题:
 *
内 容:
Email:  *
验证码:   刷新

Items in IR are protected by copyright, with all rights reserved, unless otherwise indicated.

 

 

Valid XHTML 1.0!
Copyright © 2007-2016  中国科学院沈阳自动化研究所 - Feedback
Powered by CSpace