基于Hough变换的直线提取方法及改进
|
摘要
随着信息时代的到来,图像特征提取成为三维城市建模的关键问题之一,同时也是计算机视觉和图像理解领域的研究前沿,而在特征提取过程中,直线提取方法尤为重要。鉴于Hough变换直线提取具有几何解析性简单、容错性和鲁棒性好,能够较好地处理局部遮挡、覆盖等优点,本文把基于Hough变换的直线提取作为研究重点。首先,对国内外Hough变换直线提取方法的产生及研究现状进行了概述,然后,利用Canny算子和Freeman链码对该方法进行改进,并对优化后的Hough变换直线提取方法进行了全面的实验与分析。结果表明,该方法取得了更加可靠的直线提取结果。
With the coming of the information age,the extraction of image features has become one of the key problems in 3 D urban modeling. It has also become a research frontier in the field of computer vision and image comprehension. In the process of feature extraction,the line extraction method is especially important. The line extraction of Hough transform has the advantages of simple geometric resolution,high fault tolerance and good robustness. It can also better deal with local occlusion and coverage. In this paper,the line extraction based on Hough transform is used as the research focus. Firstly,this paper gives a comprehensive overview of the generation and research status of Hough transform line extraction methods at home and abroad. Then the Canny operator and the Freeman chain code are used to improve the method,and some experiments and analysis are carried out for the optimized Hough transform. Experiment results show that they have been more reliable than before.
With the coming of the information age,the extraction of image features has become one of the key problems in 3 D urban modeling. It has also become a research frontier in the field of computer vision and image comprehension. In the process of feature extraction,the line extraction method is especially important. The line extraction of Hough transform has the advantages of simple geometric resolution,high fault tolerance and good robustness. It can also better deal with local occlusion and coverage. In this paper,the line extraction based on Hough transform is used as the research focus. Firstly,this paper gives a comprehensive overview of the generation and research status of Hough transform line extraction methods at home and abroad. Then the Canny operator and the Freeman chain code are used to improve the method,and some experiments and analysis are carried out for the optimized Hough transform. Experiment results show that they have been more reliable than before.
引文
[1]康文静,丁雪梅,崔继文,等.基于改进Hough变换的直线图形快速提取方法[J].光电工程,2007,34(3):105-117.
[2]潘建平,邬明权.基于数学形态的道路提取[J].计算机工程与应用,2008,44(11):232-233.
[3]王义敏,秦永元.基于区域生长的SAR图像目标检测方法研究[J].计算机应用,2009,29(1):45-46.
[4]朱昌盛,周伟,关键.基于平行线对检测的SAR图像主干道提取算法[J].中国图象图形学报,2011,16(10):1 908-1 917.
[5]许华荣,王晓栋,方遒.基于B样条曲线模型的结构化道路检测算法[J].自动化学报,2011,37(3):270-275.
[6]邢坤,何红艳,岳春宇,等.遥感图像中基于Hough变换的直线提取方法[J].航天返回与遥感,2015,36(1):87-94.
[7]刘春阁.基于自适应中值滤波和改进Hough变换的直线提取[J].测绘科学,2011,36(6):267-269.
[8]宋晓宇,郭寒冰,袁帅,等.基于自适应阈值区间的动态采样Hough变换直线检测算法[J].沈阳建筑大学学报,2014,30(5):945-952.
[9]李明磊,李广云,王力,等.3D Hough Transform在激光点云特征提取中的应用[J].测绘通报,2015(2):29-33.
[10]徐南,周绍光.基于图像分块和线段投票的遥感道路边缘线提取[J].国土资源遥感,2015,27(1):55-61.
[11]VaddiRS,Boggavarapu LNP,Vankayalapati HD.Contour detection using Freeman chain code and approximation methods for the real-time object[J].Asian Iournal of Computer Science and Information Technology,2011,1(1):15-1
[12]张国英,程益钰,朱红.基于改进Hough变换的线性目标检测[J].计算机工程与设计,2014,35(2):536-540.
[2]潘建平,邬明权.基于数学形态的道路提取[J].计算机工程与应用,2008,44(11):232-233.
[3]王义敏,秦永元.基于区域生长的SAR图像目标检测方法研究[J].计算机应用,2009,29(1):45-46.
[4]朱昌盛,周伟,关键.基于平行线对检测的SAR图像主干道提取算法[J].中国图象图形学报,2011,16(10):1 908-1 917.
[5]许华荣,王晓栋,方遒.基于B样条曲线模型的结构化道路检测算法[J].自动化学报,2011,37(3):270-275.
[6]邢坤,何红艳,岳春宇,等.遥感图像中基于Hough变换的直线提取方法[J].航天返回与遥感,2015,36(1):87-94.
[7]刘春阁.基于自适应中值滤波和改进Hough变换的直线提取[J].测绘科学,2011,36(6):267-269.
[8]宋晓宇,郭寒冰,袁帅,等.基于自适应阈值区间的动态采样Hough变换直线检测算法[J].沈阳建筑大学学报,2014,30(5):945-952.
[9]李明磊,李广云,王力,等.3D Hough Transform在激光点云特征提取中的应用[J].测绘通报,2015(2):29-33.
[10]徐南,周绍光.基于图像分块和线段投票的遥感道路边缘线提取[J].国土资源遥感,2015,27(1):55-61.
[11]VaddiRS,Boggavarapu LNP,Vankayalapati HD.Contour detection using Freeman chain code and approximation methods for the real-time object[J].Asian Iournal of Computer Science and Information Technology,2011,1(1):15-1
[12]张国英,程益钰,朱红.基于改进Hough变换的线性目标检测[J].计算机工程与设计,2014,35(2):536-540.