Table of Contents Author Guidelines Submit a Manuscript
Mathematical Problems in Engineering
Volume 2018, Article ID 7824279, 11 pages
https://doi.org/10.1155/2018/7824279
Research Article

Family of -Ary Univariate Subdivision Schemes Generated by Laurent Polynomial

Department of Mathematics, The Islamia University of Bahawalpur, Bahawalpur, Pakistan

Correspondence should be addressed to Ghulam Mustafa; kp.ude.bui@afatsum.maluhg

Received 13 September 2017; Revised 27 February 2018; Accepted 2 April 2018; Published 13 May 2018

Academic Editor: Alessandro Gasparetto

Copyright © 2018 Muhammad Asghar and Ghulam Mustafa. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

The generalized symbols of the family of -ary univariate stationary and nonstationary parametric subdivision schemes have been presented. These schemes are the new version of Lane-Riesenfeld algorithms. Comparison shows that our proposed family has higher continuity and generation degree comparative to the existing subdivision schemes. It is observed that many existing binary and ternary schemes are the special cases of our schemes. The analysis of proposed family of subdivision schemes is also presented in this paper.

1. Introduction

Subdivision schemes are powerful tools in CAGD for generation of smooth curves and surfaces. Subdivision schemes are popular in many practical applications such as multiresolution modeling and character animation.

The curve generation in computer graphics especially by subdivision schemes has applications in signal compression [1]. It has been shown that the subdivision schemes are suitable algorithm for compressing both regular and fractal-like signals. Herley [2] used subdivision schemes to derive necessary and sufficient conditions under which a discrete-time signal can be exactly interpolated from only one out of every sample and show how designs may be carried out. Taubin [3] described a new tool for interactive free-form fair surface design. By generalizing classical discrete Fourier analysis to two-dimensional discrete surface signals- functions defined on polyhedral surfaces of arbitrary topology.

Initially Lane and Riesenfeld present an algorithm [4] for subdividing uniform B-spline schemes of order , with . After that, this algorithm was used in different variants [5]. In stationary context, Deslauriers-Dubuc schemes [6] were characterized by a symbol containing the factor . Hormann and Sabin [7] presented a family of subdivision schemes with cubic precision (with ). Ashraf et al. [8] presented a new family of subdivision schemes by using six-point variant on the Lane-Riesenfeld algorithm. In nonstationary context, Conti and Romani [5] presented general affine combination of B-spline subdivision masks and its nonstationary counterparts. Conti and Romani [9] also presented algebraic conditions on nonstationary subdivision schemes for exponential polynomial reproduction. Novara and Romani [10] offered an extension of nonstationary Lane-Riesenfeld algorithm and a nonstationary family of alternating primal/dual subdivision schemes with reproduction of .

During the last decade, much work has been published on parametric subdivision schemes. Parametric subdivision schemes are good to control the shapes in curve and surface design. Siddiqi and Rehan [11] modified 3-point binary and ternary subdivision schemes with a tension parameter which generate a family of limiting curves for certain range of tension parameter. Using same technique, they also presented an improved four point scheme [12]. Mustafa et al. [13] presented a general formula for odd point ternary approximating subdivision schemes with a shape parameter. Ghaffar et al. [14] proposed a 3-point scheme for any arity. They also presented 4-point, -ary subdivision schemes with tension parameter [15]. Zheng et al. [16] presented a family of integer point binary subdivision schemes with tension parameter.

1.1. Motivation

Lane-Riesenfeld algorithm has been used for the generation of a family of binary parametric subdivision schemes [17, 18]. But till now, nobody has used Lane-Riesenfeld algorithms for the generation of higher arity parametric subdivision schemes. The benefits of higher arity subdivision schemes are as follows:(i)The rate of convergence increases with the increase of arity of the scheme.(ii)The support of the schemes decreases as arity increases.

The main objective of this work is to present univariate parametric high arity subdivision schemes. The objective has been obtained by using Lane-Riesenfeld algorithm and parametric Laurent polynomials. The highlights of proposed work are as follows:(i)The generalization of Lane-Riesenfeld algorithm to generate high arity schemes(ii)A unified way to present families of univariate stationary and nonstationary high arity parametric subdivision schemes(iii)To propose families of schemes with higher continuity and generation degree comparative to the existing schemes [7, 1719] (see Tables 3 and 4)(iv)To present families of schemes so that existing binary and ternary subdivision schemes become special cases (see Section 4.1).

The paper is organized as follows. In Section 2, we present construction of family of -ary univariate approximating subdivision schemes. Analysis of the proposed family is presented in Section 3. Comparison and special cases of our proposed family are presented in Section 4. Section 5 is for nonstationary version of Section 2. Conclusions are drawn in Section 6.

2. Construction of Algorithm

In this section, a generalized B-spline symbol for -ary scheme has been presented using the well-known Lane-Riesenfeld algorithm. This algorithm is based on the smoothing operatorand refining factorNow by applying the smoothing operator , times, after one application of the refining operator , we getIts simplest form iswhich is the general symbol of th degree polynomial B-spline. The symbol defined in (4) is able to generate

Laurent polynomial (symbol) for odd arity (i.e., is odd integer) is defined asLaurent polynomial (symbol) for even arity (i.e is even integer) is defined asThe general symbols for odd and even arity schemes can be defined asrespectively. From (7) and (8), we get -ary subdivision schemes with tension parameter . The members of the families can be easily obtained corresponding to the values of and . Table 1 gives the complexity and mask of the subdivision schemes.

Table 1: Here we present arity, complexity, and mask of the schemes corresponding to different values of ; here shows complexity of the schemes (i.e., 2-, 3--point schemes), while , , and stand for arity and mask of the schemes, respectively.

3. Analysis of and Schemes

In this section, we will present the complete analysis of a family of -ary subdivision schemes. We present continuity, Hölder regularity, generation degree, and reproduction degree of proposed family of schemes.

3.1. Convergence and Smoothness Analysis

Continuity is an important property of subdivision schemes. Continuity of a subdivision scheme refers to the differentiability of the limit curve produced by subdivision process. A high continuity subdivision scheme gives more smooth limit curve. We use Laurent polynomial method [20] to calculate integer class continuity of the and -schemes. Hölder regularity is an extension of convergence and continuity which gives more information about any scheme. Lower and upper bounds on Hölder continuity are calculated by using Riouls method [21]. Moreover, exact upper bounds on Hölder continuity can also be derived by using Floater and Muntingh algorithm [22].

Theorem 1. The family of -ary subdivision schemes corresponding to the symbols and is continuous.

Proof. From (8), we haveThis implieswith , where .In binary case, the condition for is . For all other even arity subdivision schemes the condition for is . Therefore, by [20], if , then is contractive and converges. If is convergent then scheme corresponding to is continuous.
Similarly from (7), we haveThis implieswith , where . The condition for is . Therefore, by [20], if , then is contractive and converges. If is convergent then scheme corresponding to is continuous.

Theorem 2. The lower bound of Hölder regularity of a family of -ary subdivision schemes corresponding to the symbol is computed as , where is defined as and . The value of depends on .

Proof. From (9), we have, , and are the matrices with elements . By [21] the lower bound of Hölder regularity is given by , where is the joint spectral radius of the matrices ; that is, . For bounds on Hölder regularity we calculate Since is bounded from below by the spectral radii and from above by the norm of the metrics , then This implies that depends on the value of . The lower bound Hölder regularity of the scheme corresponding to the symbol is This completes the proof.

Remark 3. The upper bound of Hölder regularity of the families of binary and quaternary schemes corresponding to symbols and is and , respectively, where is defined as

Remark 4. Similarly, we can easily compute the lower and upper bounds of a family of -ary subdivision schemes corresponding to the symbol .

3.2. Generation and Reproduction Analysis

The subdivision scheme with symbol reproduces polynomials of degree with respect to the parameterizations if and only ifPolynomial reproduction of degree requires polynomial generation of degree [20].

Theorem 5. Generation degree of family of -ary subdivision schemes corresponding to the symbols and is .

Proof. The Laurent polynomial of family of even ary subdivision schemes defined in (9) can be written as where .
Similarly the Laurent polynomial of family of odd ary subdivision schemes defined in (12) can be written as where .
Hence generation degree is .

Theorem 6. The family of binary schemes corresponding to the symbol reproduces polynomial of degree with respect to the dual parameterizations for odd values of and primal parametrization for even values of .

Proof. By taking the first derivative of (9) and putting , we get . This implies that , so the scheme corresponding to the symbol has dual parametrization for odd values of and primal parametrization for even values of . We can easily verify that the first and second derivatives of at are equal to 0. Further we can also verify (17) for and . This completes the proof.

Remark 7. The family of ternary schemes corresponding to the symbol reproduces polynomial of degree with respect to the dual parameterizations with .

Remark 8. The family of quaternary schemes corresponding to the symbol reproduces polynomial of degree with respect to the dual parameterizations for odd values of and primal parametrization for even values of with .

Remark 9. The family of quinary schemes corresponding to the symbol reproduces polynomial of degree with respect to the dual parameterizations with .

Similarly, we can compute the degree of reproduction of a family of -ary subdivision schemes corresponding to the symbols and . In Table 2, we briefly present the above analysis of the schemes and .

Table 2: Here , , , , and are continuity, generation degree, reproduction degree, shift parameter, and parametrization, respectively.
Table 3: Here we the present comparison of the family of binary schemes. Here N.E, , , and OA are number of entries in mask, continuity analysis, generation degree, support and approximation order of subdivision schemes.
Table 4: Here we present the comparison of the family of ternary schemes. Here , , , and are complexity, continuity analysis, support size, and generation degree of subdivision schemes, respectively.

4. Comparison and Special Cases of and

In this section, we will present the special cases, comparison, and applications of the schemes.

4.1. Special Cases of the Schemes and

It is observed that many existing binary and ternary subdivision schemes are the special cases of our proposed and schemes.

In binary case i.e., , the following existing schemes are the special cases:(i)If , we get B-spline of th degree [4].(ii)If we take , it gives Hormann and Sabin’s family [7].(iii)If we take and , we get the general formula of integer point binary approximating subdivision schemes [16].(iv)For and , we have the mask of well-known 2-point Chaikin’s scheme described in [23].(v)If we take and , we get a 3-point binary approximate subdivision scheme described in [11].(vi)For and , we get a 3-point binary scheme [11].(vii)By setting and , we get a 3-point binary scheme [14].(viii)For and , we get a 3-point binary scheme [24].(ix)For and , we get a 3-point binary scheme [25].(x)For and , we get a 3-point binary scheme [26].(xi)By setting and , we get a 3-point binary scheme [27].(xii)For and , we get a 3-point binary scheme [28].(xiii)By setting and , we get a 3-point binary scheme [29].(xiv)For and , we get a 4-point binary scheme [12].(xv)If we take , we get a 4-point binary scheme [15].(xvi)If we take and , we get a 4-point binary scheme [30].(xvii)For and , we get a 4-point binary scheme [31].(xviii)If we take and , we get a 4-point binary scheme [19].(xix)For and , we get a 5-point binary relaxation scheme [32].(xx)By setting and , we get a 5-point binary relaxation scheme [33].(xxi)After substituting into (8), we obtain a general symbol for stationary quasi-splines [34].

In Ternary case (i.e., ), the following existing schemes are the special cases:(i)By setting and , we get odd point ternary approximating subdivision scheme described in [13].(ii)For and , we get a modified ternary 3-point scheme described in [11].(iii)If we take and , we get a 3-point ternary scheme [35].(iv)By setting and , we get a 3-point ternary scheme [14].

4.2. Application and Comparison

Aim of this subsection is to present the comparison of , , and existing subdivision schemes. In Table 3, we present the comparison of proposed family of binary schemes with existing schemes having the same number of entries in mask and same support size. In this table, the comparison of continuity, generation degree, support, and approximation order of the subdivision schemes is presented. We see that continuity and generation degree of our proposed family of binary schemes are higher comparative to the existing subdivision schemes. Similarly comparison of ternary subdivision schemes is presented in Table 4. Visual performances of our proposed family members at different values of are presented in Figure 1. We also observed that the maximum smoothness is obtained at . In Figure 2, we show that our proposed scheme gives better smoothness comparative to existing schemes of [18] and of [36].

Figure 1: Limiting close curves generated by different members of the family of -ary schemes at , , , and .
Figure 2: , , and present comparison between limit curves for close polygons produced by schemes corresponding to , of [18] and of [36] at .

5. Nonstationary Algorithm

Let us define where

We also define Remember that if then ; similarly we haveIn nonstationary context, we define the Laurent polynomial for odd arity asand Laurent polynomial for even arity aswhereThen general symbols for odd arity and even arity subdivision schemes can be defined asrespectively.

(27) and (28) are the general symbols of family of -ary subdivision schemes with tension parameter. By substituting the different values of and , we get the symbol of family members of -ary nonstationary subdivision schemes.

5.1. Family of Binary Nonstationary Schemes

In this subsection, we will present the nonstationary family of binary subdivision schemes. After substituting in (28), we get family of binary nonstationary subdivision schemes. The general symbol of binary nonstationary subdivision schemes is given byWe get the family members of a family of nonstationary subdivision schemes at different values of .

For , we get a nonstationary dual 3-point schemeFor , we get a relaxed 3-point nonstationary primal schemeFor , we get a 4-point dual nonstationary scheme Similarly, we can generate family of ternary, quaternary, and -ary subdivision schemes. Figure 3 represents the application of nonstationary scheme at different values of .

Figure 3: Reproduction of conics by scheme corresponding to .

Lemma 10. The 3-point nonstationary scheme defined in (30) is asymptotically equivalent to the scheme

Proof. We can easily verify the above result using (23).

Similarly, we have the following lemmas.

Lemma 11. The relaxed 3-point nonstationary scheme defined in (31) is asymptotically equivalent to the scheme

Lemma 12. The 4-point nonstationary scheme defined in (32) is asymptotically equivalent to the scheme

Lemma 13. The Laurent polynomials of even and odd ary nonstationary schemes are asymptotically equivalent to the Laurent polynomials of even and odd ary stationary schemes, respectively.

Proof. By (23) and (26), ; hence the proof is completed.

Lemma 14. The family of -ary nonstationary subdivision schemes corresponding to the nonstationary symbols and are asymptotically equivalent to the family of -ary stationary subdivision schemes corresponding to the symbols and , respectively.

Proof. We can easily verify the above result using (27), (28), and Lemma 13.

6. Conclusion

In this paper, we have presented a simplest way to construct the -ary univariate stationary and nonstationary parametric subdivision schemes. Our proposed families of schemes have good properties comparative to the exiting subdivision schemes. It is also observed that many existing binary and ternary subdivision schemes are the special cases of our proposed families. We also present the analysis of families of subdivision schemes. Purposed algorithms are the extension of the well-known Lane-Riesenfeld algorithm in both stationary and nonstationary context.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Acknowledgments

This work is supported by the National Research Program for Universities (NRPU) (P. no. 3183), Pakistan.

References

  1. G. Mustafa, F. Chen, and Z. Huang, “Ternary wavelets and their applications to signal compression,” International Journal of Applied Mathematics and Computer Science, vol. 14, no. 2, pp. 233–240, 2004. View at Google Scholar
  2. C. Herley, “Exact Interpolation and Iterative Subdivision Schemes,” IEEE Transactions on Signal Processing, vol. 43, no. 6, pp. 1348–1359, 1995. View at Publisher · View at Google Scholar · View at Scopus
  3. G. Taubin, “Signal processing approach to fair surface design,” in Proceedings of the 22nd Annual Conference on Computer Graphics and Interactive Techniques, pp. 351–358, Not Known, 1995.
  4. J. M. Lane and R. F. Riesenfeld, “A theoretical development for the computer generation and display of piecewise polynomial surfaces,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 2, no. 1, pp. 35–46, 1980. View at Google Scholar · View at Scopus
  5. C. Conti and L. Romani, “Affine combination of B-spline subdivision masks and its non-stationary counterparts,” BIT Numerical Mathematics, vol. 50, no. 2, pp. 269–299, 2010. View at Publisher · View at Google Scholar · View at Scopus
  6. G. Deslauriers and S. Dubuc, “Symmetric iterative interpolation processes,” Constructive Approximation, vol. 5, no. 1, pp. 49–68, 1989. View at Publisher · View at Google Scholar · View at MathSciNet · View at Scopus
  7. K. Hormann and M. A. Sabin, “A family of subdivision schemes with cubic precision,” Computer Aided Geometric Design, vol. 25, no. 1, pp. 41–52, 2008. View at Publisher · View at Google Scholar · View at MathSciNet · View at Scopus
  8. P. Ashraf, G. Mustafa, and J. Deng, “A six-point variant on the Lane-Riesenfeld algorithm,” Journal of Applied Mathematics, vol. 2014, Article ID 628285, 7 pages, 2014. View at Publisher · View at Google Scholar · View at MathSciNet
  9. C. Conti and L. Romani, “Algebraic conditions on non-stationary subdivision symbols for exponential polynomial reproduction,” Journal of Computational and Applied Mathematics, vol. 236, no. 4, pp. 543–556, 2011. View at Publisher · View at Google Scholar · View at Scopus
  10. P. Novara and L. Romani, “Building blocks for designing arbitrarily smooth subdivision schemes with conic precision,” Journal of Computational and Applied Mathematics, vol. 279, pp. 67–79, 2015. View at Publisher · View at Google Scholar · View at Scopus
  11. S. S. Siddiqi and K. Rehan, “Modified form of binary and ternary 3-point subdivision schemes,” Applied Mathematics and Computation, vol. 216, no. 3, pp. 970–982, 2010. View at Publisher · View at Google Scholar · View at Scopus
  12. S. S. Siddiqi and K. Rehan, “Improved binary four point subdivision scheme and new corner cutting scheme,” Computers & Mathematics with Applications, vol. 59, no. 8, pp. 2647–2657, 2010. View at Publisher · View at Google Scholar · View at Scopus
  13. G. Mustafa, A. Ghaffar, and F. Khan, “The odd-point ternary approximating schemes,” American Journal of Computational Mathematics, vol. 1, no. 2, pp. 111–118, 2011. View at Publisher · View at Google Scholar
  14. A. Ghaffar, G. Mustafa, and K. Qin, “Unification and application of 3-point approximating subdivision schemes of varying arity,” Open Journal of Applied Sciences, vol. 2, no. 4, pp. 48–52, 2012. View at Publisher · View at Google Scholar
  15. A. Ghaffar, G. Mustafa, and K. Qin, “The 4-point α-ary approximating subdivision scheme,” Open Journal of Applied Sciences, vol. 03, no. 01, pp. 106–111, 2013. View at Publisher · View at Google Scholar
  16. H. Zheng, S. Huang, F. Guo, and G. Peng, “Integer-point binary approximating subdivision schemes,” Journal of Information and Computational Science, vol. 11, no. 10, pp. 3387–3398, 2014. View at Publisher · View at Google Scholar · View at Scopus
  17. N. Dyn, A. Heard, K. Hormann, and N. Sharon, “Univariate subdivision schemes for noisy data with geometric applications,” Computer Aided Geometric Design, vol. 37, article no. 1495, pp. 85–104, 2015. View at Publisher · View at Google Scholar · View at Scopus
  18. L. Romani, “A Chaikin-based variant of Lane-Riesenfeld algorithm and its non-tensor product extension,” Computer Aided Geometric Design, vol. 32, pp. 22–49, 2015. View at Publisher · View at Google Scholar · View at MathSciNet · View at Scopus
  19. S. S. Siddiqi and N. Ahmad, “An approximating C4 stationary subdivision scheme,” European Journal of Scientific Research, vol. 15, no. 1, pp. 97–102, 2006. View at Google Scholar · View at Scopus
  20. K. Hormann, “Subdivision schemes for geometric modelling: a hands-on tutorial,” in Dolomites Research Notes on Approximation, University of Verona, Verona, Italy, 2012. View at Google Scholar
  21. O. Rioul, “Simple regularity criteria for subdivision schemes,” SIAM Journal on Mathematical Analysis, vol. 23, no. 6, pp. 1544–1576, 1992. View at Publisher · View at Google Scholar
  22. M. S. Floater and G. Muntingh, Exact Regularity of Symmetric Univariate Subdivision Schemes, Geometry seminar, Centre of Mathematics for Applications, University of Oslo, Oslo, Norway, 2012.
  23. G. M. Chaikin, “An algorithm for high-speed curve generation,” Computer Graphics and Image Processing, vol. 3, no. 4, pp. 346–349, 1974. View at Publisher · View at Google Scholar
  24. S. S. Siddiqi and N. Ahmad, “A new three-point approximating C2 subdivision scheme,” Applied Mathematics Letters, vol. 20, no. 6, pp. 707–711, 2007. View at Publisher · View at Google Scholar · View at Scopus
  25. G. Mustafa, A. Ghaffar, and M. Bari, “(2n-1)-point binary approximating scheme,” in Proceedings of the 8th International Conference on Digital Information Management, ICDIM '13, pp. 363–368, September 2013. View at Publisher · View at Google Scholar · View at Scopus
  26. M. F. Hassan and N. A. Dodgson, “Ternary and three-point univariate subdivision schemes,” Curve and Surface Fitting: Saint-Malo, vol. 2002, pp. 199–208, 2002. View at Google Scholar
  27. G. Mustafa, F. Khan, and A. Ghaffar, “The m-point approximating subdivision scheme,” Lobachevskii Journal of Mathematics, vol. 30, no. 2, pp. 138–145, 2009. View at Publisher · View at Google Scholar · View at Scopus
  28. S. Daniel and P. Shunmugaraj, “Three point stationary and non-stationary subdivision schemes,” in Proceedings of the 3rd International Conference on Geometric Modeling and Imaging, GMAI '08, vol. 2008, pp. 3–8, July 2008. View at Publisher · View at Google Scholar · View at Scopus
  29. G. Mustafa and A. H. Randhawa, “Complete analysis of 3-point binary subdivision scheme,” Journal of Pure and Applied Sciences, vol. 24-33, pp. 21–31, 2014. View at Google Scholar
  30. L. Zhang, Z. Luo, and W. Liu, “An approximating C3 stationary subdivision scheme,” Journal of Information Computational Science, vol. 1, no. 1, pp. 157–161, 2004. View at Google Scholar
  31. N. Dyn, M. S. Floater, and K. Hormann, “A C2 four-point subdivision scheme with fourth order accuracy and its extensions,” Analysis, vol. 1, no. 128, 12 pages, 2005. View at Google Scholar
  32. H. Cao and J. Tan, “A binary five-point relaxation subdivision scheme,” Journal of Information and Computational Science, vol. 10, no. 18, pp. 5903–5910, 2013. View at Publisher · View at Google Scholar · View at Scopus
  33. J. Tan, Y. Yao, H. Cao, and L. Zhang, “Convexity preservation of five-point binary subdivision scheme with a parameter,” Applied Mathematics and Computation, vol. 245, pp. 279–288, 2014. View at Publisher · View at Google Scholar · View at Scopus
  34. B. Jeong, Y. J. Lee, and J. Yoon, “A family of non-stationary subdivision schemes reproducing exponential polynomials,” Journal of Mathematical Analysis and Applications, vol. 402, no. 1, pp. 207–219, 2013. View at Publisher · View at Google Scholar · View at Scopus
  35. M. Aslam, G. Mustafa, and A. Ghaffar, “(2n-1)-point ternary approximating and interpolating subdivision schemes,” Journal of Applied Mathematics, vol. 2011, Article ID 832630, 12 pages, 2011. View at Publisher · View at Google Scholar · View at MathSciNet
  36. G. Mustafa and R. Hameed, “Families of univariate and bivariate subdivision schemes originated from quartic B-spline,” Advances in Computational Mathematics, pp. 1–31, 2017. View at Google Scholar
  37. M. F. Hassan, “Further analysis of ternary and 3-point univariate subdivision schemes,” Tech. Rep. 599, University of Cambridge Computer Laboratory, Cambridge, UK, 2004. View at Google Scholar
  38. A. Levin, “Polynomial generation and quasi-interpolation in stationary non-uniform subdivision,” Computer Aided Geometric Design, vol. 20, no. 1, pp. 41–60, 2003. View at Publisher · View at Google Scholar · View at Scopus