参数优化模型的正交逼近算法
doi: 10.3724/SP.J.1146.2008.01585
Orthogonal Approximation Algorithm for Parameter Model Optimization
-
摘要: 在图像拼接中,该文将图像配准转化成一个无约束最优化问题,运用2次型理论提出一种新的参数优化模型算法正交方向逼近法。该算法根据参数矩阵最优函数特有的性质,无须计算函数的导数或梯度,仅须计算函数值即可。其优点是克服了以往算法所要求占用大量计算机内存的缺点,仅用一个矩阵存储数字信息。通过仿真实验证明:该算法收敛速度快,结果精度高,具有较高的实用价值。
-
关键词:
- 图像拼接; 图像配准; 凹凸性; 最优化
Abstract: Image registration is transformed an unconstrained optimization problem in image mosaic. Parameter optimization model algorithm is proposed by means of quadratic form theory-- orthogonal approximation algorithm. According to properties of best-function, a direct method is adopted without calculating derivative or gradient, only calculating function values. The advantages are eliminating of taking up large computer memory, only using a matrix to store information. Simulation experiments show the algorithm has advantages of fast convergence, result precisely and good practicality. -
赵向阳, 杜立民. 一种全自动自动稳健的图像自动拼接融合算法[J]. 中国图像图形学报, 2004, 9(4): 417-422.[2]Brown M and Lowe D G. Recognising panoramas[C].Proceedings of IEEE International Conference on ComputerVision, Nice, France, 2003, 2: 1218-1255.[3]李艳丽, 向辉. 稳健的球面全景图全自动生成算法[J]. 计算机辅助设计与图形学报, 2007, 19(11): 1393-1398.[4]Hartley R I. In defence of the 8-point algorithm[C].Proceedings of IEEE International Conference on ComputerVision, Cambridge, Britain, 1995: 1064-1070.[5]杨敏. 基于Levenberg-Marquardt 算法的图像拼图研究[J]. 电子工程师, 2007, 33(10): 43-45.[6]北京大学数学系几何与代数教研室代数小组. 高等代数[M].北京: 高等教育出版社, 2000: 210-241.[7]江苏师范学院数学系《解析几何》编写组. 解析几何[M]. 北京: 高等教育出版社, 1988: 244-265.[8]韩凌燕. 二次型化简二次曲线方法的探究[J]. 山东科学, 2008,21(2): 65-67.[9]Lowe D G. Distinctive image features from scale-invariantkeypoints[J].International Journal of Computer Vision.2004,60(2):91-110[10]Chum O, Matas J, and Kittler J. Locally OptimizedRANSAC[J].DAGM Symposium on Pattern Recognition,LNCS.2003, 2781:236-243
计量
- 文章访问数: 3248
- HTML全文浏览量: 78
- PDF下载量: 640
- 被引次数: 0