一种基于隐式多项式曲线的形状自适应描述方法

doi:10.3969/j.iss.0372-2012-2014.03.013

电子学报 ›› 2014, Vol. 42 ›› Issue (3): 505-511.DOI: 10.3969/j.iss.0372-2012-2014.03.013

一种基于隐式多项式曲线的形状自适应描述方法

吴刚

南京财经大学电子商务重点实验室, 江苏南京 210003

收稿日期:2012-02-14 修回日期:2013-07-21 出版日期:2014-03-25
- 作者简介:
- 吴刚男，1969年6月出生江苏邳州，博士.主要从事模式识别和计算机视觉方面的研究工作.E-mail：wugang69@gmail.com
- 基金资助:
- 江苏省自然科学基金 (No.BK2009349）

An Adaptive Shape Description Method Based on Implicit Polynomial Curves

WU Gang

Key Lab of Electronic Business, Nanjing University of Finance and Economic, Nanjing, Jiangsu 210003, China

Received:2012-02-14 Revised:2013-07-21 Online:2014-03-25 Published:2014-03-25
- Supported by:
- National Natural Science Foundation of Jiangsu Province, China (No.BK2009349)

摘要/Abstract

摘要： 隐式多项式曲线对目标形状的线性描述算法需要构造隐式多项式曲线的代数残差和梯度残差，但是如何加权平衡这两种残差是一个困难问题.提出了一种平衡这两种残差的自适应权重确定算法.根据隐式多项式曲线稳定性要求，构造了隐式多项式曲线系数向量范数和权重的关系式，进而计算出极小值点并作为最优权重，实现了隐式多项式曲线的自适应描述.从MPEG-7标准测试集中选取不同形状的图形进行描述，实验结果表明，根据自适应权重获取的隐式多项式曲线不仅保证了描述的精确性，而且十分稳定，没有出现多余的曲线.

关键词: 隐式多项式曲线, 形状描述, 自适应权重

Abstract: The residual sum of squares of implicit polynomial (IP) and its gradients are constructed as objective function that can be minimized to obtain the fitted IP curve for representing a given shape.However,how to balance the two residual sum of squares using a weight factor is a difficult problem.The adaptive weight algorithm was proposed to determine the weight factor.Firstly,the function formula for the coefficient vector norm of IP curve and the weight were constructed according to the stability of IP curve.Secondly,the minimizer of the the function was computed,and used as the weight.Lastly,the IP curve can be obtained by minimizing the objective function with weight.The performance of the proposed method was evaluated on the standard MPEG-7 shape database,and the experimental results show that the IP curve obtained by proposed algorithm can represent the object shapes both accurately and stably.

Key words: implicit polynomial curve, shape description, adaptive weight

中图分类号:

TP391.41

吴刚. 一种基于隐式多项式曲线的形状自适应描述方法[J]. 电子学报, 2014, 42(3): 505-511.

WU Gang . An Adaptive Shape Description Method Based on Implicit Polynomial Curves[J]. Acta Electronica Sinica, 2014, 42(3): 505-511.

参考文献

[1] G Taubin, F Cukirman, S Sullivan.Parameterized families of polynomials for bounded algebraic curve and surface fitting[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 1994, 16(3):286-303.

[2] 吴刚, 李道伦.基于隐含多项式曲线仿射不变量的目标识别[J].电子学报, 2004, 32(12):1987-1991. Wu Gang, Li Dao-lun.Object recognition based on affined invariants in implicit polynomial curves[J].Acta Electronica Sinica.2004, 32(12):1987-1991.(in Chinese)

[3] T Tasdizen.Robust and Repeatable Fitting of Implicit Polynomial Curves to Point Data Sets and to Intensity Images[D].Rhode Island, USA:Brown University, 2001.

[4] D Keren, D Cooper.Describing complicated objects by implicit polynomial[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 1994, 16(1):38-53.

[5] S Sullivan, L Sandford, JPonce.Using geometric distance fits for 3D object modeling and recognition[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 1994, 16(12):1183-1196.

[6] D Keren, Craig Gotsman.Fitting curves and surfaces with constrained implicit polynomials[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 1999, 21(1):31-41.

[7] M Blane, Z lei.The3L algorithm for fitting implicit polynomial curves and surfaces to data[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 2000, 22(3):298-313.

[8] T Tasdizen, J Tarel.Improving the stability of algebraic curves for application[J].IEEE Trans Image Processing, 2000, 9(3):405-416.

[9] A Heizer, M Barzohar, D malah.Stable fitting of 2D curves and 3D surfaces by implicit polynomials[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 2004, 26(10):1283-1294.

[10] Bo Zheng, Jun Takamatsu, Katsushi lkeuchi.An adaptive and stable method for fitting implicit polynomial curves and surfaces[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010, 32(3):561-567.

[11] D Zhang, G Lu.Study and evaluation of different fourier method for image retrieval[J].Image and Vision Computing, 2005, 23(1):33-49.

[12] 吴刚.隐含多项式曲线曲面拟合次数的确定研究[J].计算机研究与发展, 2007, 44(1):148-153. Wu Gang.Research on degree of fitting implicit polynomial curves and surfaces[J].Journal of Computer Research and Development, 2007, 44(1):148-153.(in Chinese)

一种基于隐式多项式曲线的形状自适应描述方法

An Adaptive Shape Description Method Based on Implicit Polynomial Curves

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 10

编辑推荐

Metrics

[1]	褚鼎立, 陈红, 王旭光. 基于自适应权重和模拟退火的鲸鱼优化算法[J]. 电子学报, 2019, 47(5): 992-999.
[2]	田泽宇, 门朝光, 汤亚楠. 基于形状及空间关系的场景相似性检索[J]. 电子学报, 2016, 44(8): 1892-1898.
[3]	齐现英, 刘伯强, 徐建伟. 基于不确定性信息融合的高密度椒盐噪声降噪方法[J]. 电子学报, 2016, 44(4): 878-885.
[4]	王斌. 一种基于多尺度拱高形状描述的图像检索方法[J]. 电子学报, 2013, 41(9): 1821-1825.
[5]	唐贵进, 朱秀昌, 刘天亮. 一种基于局部可信视差的立体图像误码掩盖算法[J]. 电子学报, 2013, 41(4): 781-786.
[6]	王斌;. 一种不变的基于傅立叶变换的区域形状描述子[J]. 电子学报, 2012, 40(1): 84-88.
[7]	汤进;陈展展;罗斌;孙登第. 基于复杂网络和最优子序列双射的形状描述与匹配[J]. 电子学报, 2011, 39(8): 1757-1765.
[8]	王斌;. 一种用于形状描述的拱高半径复函数[J]. 电子学报, 2011, 39(4): 831-836.
[9]	周明全;樊亚春;耿国华. 一种基于空间对称变换的三维模型形状描述方法[J]. 电子学报, 2010, 38(4): 853-0859.
[10]	王文彬;孙其博;赵新超;杨放春. 基于非均衡变异离散粒子群算法的QoS 全局最优Web服务选择方法[J]. 电子学报, 2010, 38(12): 2774-2779.