摘要
针对传统的基于Kruppa方程摄像机自标定算法的欠鲁棒性,首次提出将鲁棒的张量投票算法用于摄像机自标定方法中。利用基于尺度不变的SIFT算法查找并匹配出每对图像的特征点,其中待匹配图像由摄像机对同一场景从三个不同角度位置拍摄,对图像张量投票后按棒张量特征值降序排序,由此筛选得到具有鲁棒性边缘特征的前八对特征点,利用八点算法求解相应的基础矩阵和极点,根据Kruppa方程和三维重建(SFM)算法求得摄像机参数矩阵。实验结果证明,该方法具有较高标定精度,并通过加入高斯噪声的仿真实验证明该算法是一种鲁棒的摄像机自标定方法。
In order to improve robustness of the traditional camera self-calibration algorithm based on Kruppa equations,the new method of camera self-calibration based on tensor voting is first proposed. The SIFT algorithm, based on scale invariant property, is adopted to extract and match the feature points of each image, which are taken from three different angles to the same scene. The first eight feature points with robustness are figured out with tensor voting and sorting.Then the fundamental matrix and the pole points are calculated by 8 points' algorithm, and finally the parameter matrices of camera can be obtained by the Kruppa equations and the Structure From Motion(SFM)algorithm. Compared with other algorithms through the comparative experiments, this method is proven to be more accurate, meanwhile, it can be regarded as a new robust camera self-calibration algorithm by the simulation experiments with Gaussian noise.
引文
[1]孟晓桥,胡占义.摄像机自标定方法的研究与进展[J].自动化学报,2003,29(1):110-124.
[2]吴福朝.计算机视觉中的数学方法[M].北京:科学出版社,2008:59-67.
[3]Anthony W,Gerhard R.Estimating intrinsic camera parameters from the fundamental matrix using an evolutionary approach[J].EURASIP Journal on Applied Signal Processing,2004(8):1113-1124.
[4]郭秋艳,刘鹏飞,安平,等.基于遗传算法的摄像机自标定方法[J].中国图象图形学报,2006,11(11):1712-1715.
[5]郝泳涛,周薇,钟波涛.平移初值操作的基于Kruppa方程的自标定方法[J].计算机工程与应用,2009,45(22):38-40.
[6]董鸿燕,沈振康,罗军,等.感知编组综述[J].计算机工程与应用,2007,43(14):9-13.
[7]张永,武玉建.一种改进的SIFT图像特征匹配算法[J].计算机工程与应用,2014,50(9):167-169.
[8]Moreno R,Garcia M A.On improving the efficiency of tensor voting[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2011,33(11):2215-2228.
[9]邵晓芳,孙即祥.改进的张量投票算法[J].计算机辅助设计与图形学学报,2006,18(7):1028-1031.
[10]Habed A,Boufama B.Camera self-calibration from bivariate polynomials derived from Kruppa’s equations[J].Pattern Recognition,2008,41(8):2484-2492.
[11]Medioni G,Lee M S,Tang C K.A computational framework for feature extraction and segmentation[M].New York:Elsevier,2000:33-64.
[12]雷成,吴福朝,胡占义.一种新的基于Kruppa方程的摄像机自标定方法[J].计算机学报,2003,26(5):1-11.
[13]Dai Yuchao,Li Hongdong,He Mingyi.A simple priorfree method for non-rigid structure-from-motion factorization[J].International Journal of Computer Vision,2014,107(2):101-122.
[14]Falconer K,Xiao Yimin.Generalized dimensions of images of measures under Gaussian processes[J].Advances in Mathematics,2014:492-517.
[15]Zhang Z Y.A flexible new technique for camera calibration[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2000,22(11):1330-1334.
[16]邢文训.现代优化计算方法[M].北京:清华大学出版社,2007:5-10.