简宏伟,陈 震

摘 要:在基于直线光流场的三维重建研究中首先要对连续多帧单目图像序列进行处理,主要是二维图像直线的提取、跟踪及匹配。在此利用改进的Hough变换对图像进行直线提取,利用Hough变换的点线对偶性,即可将对图像序列中的特征直线的跟踪与匹配问题转化为在参数空间(ρ,θ)中对特征点运动轨迹的参数估计,可以采用卡尔曼滤波器对运动目标进行初步跟踪,预测其大致位置,再通过建立模板与图像的匹配相关系数判定搜索到的特征点的精确坐标。实验结果表明,该方法取得了较好的实验结果。

关键词:边界算子;Hough变换;卡尔曼滤波器;模板

中图分类号:TN919文献标识码:A

文章编号:1004-373X(2009)20-066-03

Extraction and Tracking Match of 2-D Image Array Straight Line

JIAN Hongwei,CHEN Zhen

(College of Automation,Nanchang Hangkong University,Nanchang,330063,China)

Abstract:During the study of reconstructing 3D based on straight line optical flow,a sequence of monocular images should be processed first.Using improved Hough transform to extract lines,and a tracking characteristic lines method which is based on Hough transform is presented.Rather than tracking lines in image space directly,the method makes use of duality theory in Hough transform in order to transform line-tracking in image space into point-tracking in Hough space.The first step designing is adopting Kalman filter to preliminary track the motion target,then accurate characteristic point location is searched by using build-up modeling board and image matching correlation modulus.Experiment result using composed image array shows that the presented method provides a good estimation of the characteristic lines tracking.

Keywords:boundary operator;Hough transform;Kalman filter;modeling board

点和线是计算机视觉中最常用到的视觉特征,通常图像特征抽象的层次越高,越有利于图像的处理与分析,为了更好地进行三维重建,非常有必要提高图像特征的抽象层次,即将点级别提高到线级别。在计算机视觉和图像处理领域,直线特征相比于点特征提供了更多场景结构信息[1],并能够很稳健地检测和跟踪感兴趣的直线特征。

1 Hough变换提取直线的改进方法

1972年Duda和Hart提出利用Hough变换提取直线,Hough变换法提取直线是一种变换域提取直线的方法[2],利用线-点的对偶性,把直线上点的坐标变换到通过该点的直线的系数域[3],利用了共线和直线相交的关系[4],使直线的提取问题转化为计数问题[5]。Hough变换在直线提取方法中拥有众多优点的同时也存在着一些缺点,如需要较大的存储空间和较长的计算时间[6,7]。……