改进的RANSAC立体匹配算法的研究
|
摘要
图像匹配的精度在很大程度上决定着三维重建的成功与否.为了提高图像匹配的精度和速度,提出了一种改进的随机抽样一致性(RANSAC)算法的图像匹配方法.该方法首先采用Harris角点检测算子提取图像的特征点,然后通过SIFT算法计算特征描述子对图像进行初匹配,最后采用改进的RANSAC算法对误匹配点进行剔除,在保证精度的前提下提高了算法的速度.主要从两个方面进行RANSAC算法的改进:采用均匀分布9点算法生成的基本矩阵代替常规方法中的单应矩阵作为模型进行计算,使得模型具有较高的鲁棒性;使用随机块选取法选择样本,保障了选点的均匀分布性并且保证了精度.实验结果表明,此方法不仅能够得到较高的精确度,而且还大幅度减少了计算量,提高了匹配速度.
The accuracy of image matching determines the success of 3 D reconstruction to a great extent.In this paper,an image matching method based on modified RANSAC algorithm is proposed to improve the precision and speed. Firstly,the feature points of the images are extracted using the Harris algorithm.Then,the image pair is matched roughly by generating SIFT feature descriptor. At last,the precision of image matching is optimized by the modified RANSAC algorithm,which improved the speed of calculation under the premise of ensuring accuracy. The RANSAC algorithm is improved from the following two aspects: uses the fundamental matrix generated by the 9 point algorithm instead of the homography matrix in the traditional RANSAC algorithm as the model,which makes the model more robust and suitable for the entire data set; the sample is selected by a random block selecting method,which ensures the uniform distribution and the accuracy. The experimental results show that this method can not only get higher matching accuracy,but also greatly reduce the computation and improve the matching speed.
The accuracy of image matching determines the success of 3 D reconstruction to a great extent.In this paper,an image matching method based on modified RANSAC algorithm is proposed to improve the precision and speed. Firstly,the feature points of the images are extracted using the Harris algorithm.Then,the image pair is matched roughly by generating SIFT feature descriptor. At last,the precision of image matching is optimized by the modified RANSAC algorithm,which improved the speed of calculation under the premise of ensuring accuracy. The RANSAC algorithm is improved from the following two aspects: uses the fundamental matrix generated by the 9 point algorithm instead of the homography matrix in the traditional RANSAC algorithm as the model,which makes the model more robust and suitable for the entire data set; the sample is selected by a random block selecting method,which ensures the uniform distribution and the accuracy. The experimental results show that this method can not only get higher matching accuracy,but also greatly reduce the computation and improve the matching speed.
引文
[1]Lowe D G,Lowe D G.Distinctive image features from scale-invariant keypoints[J].International Journal of Computer Vision,2004,60(2):91-110
[2]Bay H,Tuytelaars T,Gool L V.SURF:speeded Up robust features[J].Computer Vision&Image Understanding,2006,110(3):404-417
[3]张培耘,华希俊,夏乐春,等.基于RANSAC算法的极线约束立体视觉匹配方法研究[J].组合机床与自动化加工技术,2013(11):20-22
[4]Neto V F D C,Campos M F M.An improved methodology for image feature matching[C]∥Xxii Brazilian Symposium on Computer Graphics and Image Processing.IEEE Computer Society,2009:307-314
[5]刘毅.基于一致性随机采样的图像特征匹配鲁棒确认[J].重庆邮电大学学报:自然科学版,2010,22(3):381-385
[6]Mokhtarian F,Abbasi S.Shape similarity retrieval under affine transforms[J].Pattern Recognition,2002,35(1):31-41
[7]罗文超,刘国栋,杨海燕.SIFT和改进的RANSAC算法在图像配准中的应用[J].计算机工程与应用,2013,49(15):147-149
[8]曲天伟,安波,陈桂兰.改进的RANSAC算法在图像配准中的应用[J].计算机应用,2010,30(7):1849-1851
[9]马丽丽,曹春梅,陈金广,等.基于RANSAC的特征点匹配算法[J].计算机工程与设计,2016,37(7):1794-1797
[10]邱亚辉,李长青,崔有帧.RANSAC算法在剔除图像配准中误匹配点的应用[J].影像技术,2014,26(4):46-47
[11]王昭娜,赵西安,燕青浩,等.一种核线约束模型优化与可靠性匹配方法[J].测绘科学,2016,41(10):4-9
[12]Hartley,Richard.Multiple view geometry in computer vision/2nd ed[M].Cambridge University Press,2003:159-170,191-195
[13]Fischler M A,Bolles R C.Random sample consensus:a paradigm for model fitting with applications to image analysis and automated cartography[J].Communications of the Acm,1981,24(6):381-395
[2]Bay H,Tuytelaars T,Gool L V.SURF:speeded Up robust features[J].Computer Vision&Image Understanding,2006,110(3):404-417
[3]张培耘,华希俊,夏乐春,等.基于RANSAC算法的极线约束立体视觉匹配方法研究[J].组合机床与自动化加工技术,2013(11):20-22
[4]Neto V F D C,Campos M F M.An improved methodology for image feature matching[C]∥Xxii Brazilian Symposium on Computer Graphics and Image Processing.IEEE Computer Society,2009:307-314
[5]刘毅.基于一致性随机采样的图像特征匹配鲁棒确认[J].重庆邮电大学学报:自然科学版,2010,22(3):381-385
[6]Mokhtarian F,Abbasi S.Shape similarity retrieval under affine transforms[J].Pattern Recognition,2002,35(1):31-41
[7]罗文超,刘国栋,杨海燕.SIFT和改进的RANSAC算法在图像配准中的应用[J].计算机工程与应用,2013,49(15):147-149
[8]曲天伟,安波,陈桂兰.改进的RANSAC算法在图像配准中的应用[J].计算机应用,2010,30(7):1849-1851
[9]马丽丽,曹春梅,陈金广,等.基于RANSAC的特征点匹配算法[J].计算机工程与设计,2016,37(7):1794-1797
[10]邱亚辉,李长青,崔有帧.RANSAC算法在剔除图像配准中误匹配点的应用[J].影像技术,2014,26(4):46-47
[11]王昭娜,赵西安,燕青浩,等.一种核线约束模型优化与可靠性匹配方法[J].测绘科学,2016,41(10):4-9
[12]Hartley,Richard.Multiple view geometry in computer vision/2nd ed[M].Cambridge University Press,2003:159-170,191-195
[13]Fischler M A,Bolles R C.Random sample consensus:a paradigm for model fitting with applications to image analysis and automated cartography[J].Communications of the Acm,1981,24(6):381-395