激光扫描图像快速实时匹配的研究
|
摘要
针对传统图像匹配方法对光线、视角的变化敏感,约束条件单一,匹配效率低、误差率高等难题,提出了维激光扫描图像的快速实时匹配方法。首先构建激光扫描模型,搜寻尺度空间极值,提取图像特征点及其描述,并对候选特征点进行筛选;然后采用余弦相似度和距离匹配对图像进行快速实时匹配,获取图像间精确的匹配点对,利用积分图像进一步降低匹配误差,实现图像的快速实时匹配,最后实验结果表明,本文在外界噪声干扰、光照缺失等不利条件下,大幅度缩短匹配时间,有效降低了误匹配率,提高图像匹配的精度。
The traditional image matching method is sensitive to the change of light and angle of view. The constraint condition is single,the matching is inefficient,and the error rate is above the average. This paper presents a three-dimensional laser scanning image matching method based on real-time improvement. Firstly construct the 3D laser scanning model to search the scale space extrema,extract image feature points and description. And the candidate feature points is selected. Using cosine similarity and distance matching method to real-time matching the image and to get precise image. Using the integral image to further low matching error and realize fast real-time matching. The experiments show that the 3D laser scanning image matching method proposed,in the external noise interference and the lack of illumination conditions,to shorten the matching time,reduce to effectively the error rate and to improve the accuracy.
The traditional image matching method is sensitive to the change of light and angle of view. The constraint condition is single,the matching is inefficient,and the error rate is above the average. This paper presents a three-dimensional laser scanning image matching method based on real-time improvement. Firstly construct the 3D laser scanning model to search the scale space extrema,extract image feature points and description. And the candidate feature points is selected. Using cosine similarity and distance matching method to real-time matching the image and to get precise image. Using the integral image to further low matching error and realize fast real-time matching. The experiments show that the 3D laser scanning image matching method proposed,in the external noise interference and the lack of illumination conditions,to shorten the matching time,reduce to effectively the error rate and to improve the accuracy.
引文
[1]王红梅,张科,李言俊.图像匹配研究进展[J].计算机工程与应用,2004,40(19):42-44.
[2]汪亚明.图像匹配的鲁棒型Hausdorff方法[J].计算机辅助设计与图形学学报,2002,14(3):238-241.
[3]彭真明,张启衡,魏宇星,等.基于多特征融合的图像匹配模式[J].强激光与粒子束,2004,16(3):281-285.
[4]KE SH,WANG B L,HUANG X Y.An improved SIFT algorithm and its application in medical image registra-tion[J].Journal of Xiamen University:Natural Science,2010,49(3):354-358.
[5]郑永斌,黄新生,丰松江.SIFT和旋转不变LBP相结合的图像匹配算法[J].计算机辅助设计与图形学学报,2010,22(2):286-292.
[6]王年,范益政,鲍文霞,等.基于图割的图像匹配算法[J].电子学报,2006,34(2):232-236.
[7]KOUNALAKIS T,TRIANTAFYLLIDIS G A.3D scene’s object detection and recognition using depth layers and SIFT-based machine learning[J].3D Research,2011,3(6):1-6.
[8]刘莹,曹剑中,许朝晖,等.基于灰度相关的图像匹配算法的改进[J].应用光学,2007,28(5):536-540.
[9]孔祥玲,欧斌.激光扫描技术在隧道工程竣工测量中的应用研究[J].城市勘测,2013,(2):100-102.
[10]魏志强,黄磊,纪筱鹏.基于点特征的序列图像匹配方法研究[J].中国图象图形学报,2009,14(3):525-530.
[11]冯祖仁,吕娜,李良福,等.基于最大后验概率的图像匹配相似性指标研究[J].自动化学报,2007,33(1):1-8.
[12]梅领亮,陈德勇,郝如茜,等.基于多路FIFO的宽幅CIS扫描仪图像实时重构技术[J].电子设计工程,2016,24(14):168-171.
[13]王茜.基于ARM的平板电脑硬件平台设计[J].微型机与应用,2012,31(22):29-31.
[14]刘新平,郑秋梅,孙士明,等.计算机硬件综合实验开发平台[J].实验室研究与探索,2008,27(3):48-50.
[2]汪亚明.图像匹配的鲁棒型Hausdorff方法[J].计算机辅助设计与图形学学报,2002,14(3):238-241.
[3]彭真明,张启衡,魏宇星,等.基于多特征融合的图像匹配模式[J].强激光与粒子束,2004,16(3):281-285.
[4]KE SH,WANG B L,HUANG X Y.An improved SIFT algorithm and its application in medical image registra-tion[J].Journal of Xiamen University:Natural Science,2010,49(3):354-358.
[5]郑永斌,黄新生,丰松江.SIFT和旋转不变LBP相结合的图像匹配算法[J].计算机辅助设计与图形学学报,2010,22(2):286-292.
[6]王年,范益政,鲍文霞,等.基于图割的图像匹配算法[J].电子学报,2006,34(2):232-236.
[7]KOUNALAKIS T,TRIANTAFYLLIDIS G A.3D scene’s object detection and recognition using depth layers and SIFT-based machine learning[J].3D Research,2011,3(6):1-6.
[8]刘莹,曹剑中,许朝晖,等.基于灰度相关的图像匹配算法的改进[J].应用光学,2007,28(5):536-540.
[9]孔祥玲,欧斌.激光扫描技术在隧道工程竣工测量中的应用研究[J].城市勘测,2013,(2):100-102.
[10]魏志强,黄磊,纪筱鹏.基于点特征的序列图像匹配方法研究[J].中国图象图形学报,2009,14(3):525-530.
[11]冯祖仁,吕娜,李良福,等.基于最大后验概率的图像匹配相似性指标研究[J].自动化学报,2007,33(1):1-8.
[12]梅领亮,陈德勇,郝如茜,等.基于多路FIFO的宽幅CIS扫描仪图像实时重构技术[J].电子设计工程,2016,24(14):168-171.
[13]王茜.基于ARM的平板电脑硬件平台设计[J].微型机与应用,2012,31(22):29-31.
[14]刘新平,郑秋梅,孙士明,等.计算机硬件综合实验开发平台[J].实验室研究与探索,2008,27(3):48-50.