Inter national J our nal of Electrical and Computer Engineering (IJECE) V ol. 10, No. 2, April 2020, pp. 1648 1654 ISSN: 2088-8708, DOI: 10.11591/ijece.v10i2.pp1648-1654 r 1648 A ppr oximating offset cur v es using B ´ ezier cur v es with high accuracy Abedallah Rababah 1 , Moath J aradat 2 1 Department of Mathematical Sciences, United Arab Emirates Uni v ersity , United Arab Emirates 1,2 Department of Mathematics, Jordan Uni v ersity of Science and T echnology , Jordan Article Inf o Article history: Recei v ed Apr 14, 2019 Re vised Oct 20, 2019 Accepted Oct 30, 2019 K eyw ords: Approximation order B ´ ezier curv es Circular arc Cubic approximation High accurac y ABSTRA CT In this paper , a ne w method for the approximation of of fset curv es is presented using the idea of the parallel deri v ati v e curv es. The best uniform approximation of de gree 3 with order 6 is used to construct a method to find the approximation of the of fset curv es for B ´ ezier curv es. The proposed method is based on the best uniform approximation, and therefore; t he proposed method for constructing the of fset curv es induces better outcomes than the e xisting methods. Copyright c 2020 Institute of Advanced Engineering and Science . All rights r eserved. Corresponding A uthor: Abedallah Rababah, Department of Mathematics, Jordan Uni v ersity of Science and T echnology , 22110 Irbid, Jordan. Email: rababah@just.edu.jo and rababah@uaeu.ac.ae 1. INTR ODUCTION The of fset curv es appeared in the 19th century and are widely used in Computer Aided Design/Computer Aided Manuf actoring CAD/CAM applications, and has other applications in man y computer fields. Man y studies on the of fset approximation are carried out by man y researchers. Hoschek [1] approximated the of fset curv es using splines. Rational of fset curv es are approximatedby F arouki and Sakkalis [2] by constructing the Pythagorean-hodograph (PH) curv es. In [3], rational of fset curv es based on the quadratic approximation of the circular arc are approximated. Recently , of fset approximation curv es based on the circular arc approxima- tions are presented [4-6] yielding rational of fset approximation which are the con v olution of the unit normal v ector and the gi v en curv e. The of fset approximation in this paper is based on the best uniform approxima- tion of the circular arc and yields a polynomial of fset approximation curv e. The best uniform approxima- tion of the circular arc of de gree 3 presented in [7] where the error function is the Chebyshe v polynomial of de gree 6, see also [8-16]. . This of fset method is constructed as follo ws: gi v en a B ´ ezier curv e b ( t ) and its unit normal v ector N ( t ) which is a circular arc. Then we use the best uniform approximation of de gree 3 to approximate the unit normal v ector of the gi v en curv e. Since the best uniform approximation is of high accurac y then it is anticipated that the approximation of the normal v ector is as of high accurac y . Thereafter , a special reparametrization of the approximation to unit normal v ector N a ( t ) is carried out to ha v e the same length as the unit no r mal v ector N ( t ) . In this method one step approximation is used so the error will be less than other methods. There are three types of approximation with respect to the norm; L 1 norm, L 2 norm, and L 1 norm which is the best uniform approximation that we are using in my paper . Cubic B ´ ezier curv es are commonly used in almost all industrial companies; it is used in computer graphics, animation, modeling, CAD, CA GD, design, and man y J ournal homepage: http://ijece .iaescor e .com/inde x.php/IJECE Evaluation Warning : The document was created with Spire.PDF for Python.
Int J Elec & Comp Eng ISSN: 2088-8708 r 1649 other related fields. In these and other applications in CG and CA GD, conic sections are the most commonly used curv es in an y CAD system. The Bernstein po l ynomials are one of the most important polynomials in mathematic s. The y serv e essential tasks in numerical, approximation and B ´ ezier curv es, because the y form basis which are numerically stable. The Bernstein basis polynomials of de gree n are defined as [17-19]: B n i ( t ) = n i t i (1 t ) n i ; t 2 [0 ; 1] ; i = 0 ; 1 ; 2 ; :::; n; (1) where the binomial coef fcients are gi v en by n i = n ! i !( n i )! : The Bernst ein polynomials are used as basis for the approximation and representation of curv es and are generalized to triangular surf aces [20,21]. The Bernstein polynomials are, in particular , important for the construction of the B ´ ezier curv es that are defined as follo w . A B ´ ezier curv e of de gree n is defined by b ( t ) = n X i =0 b i B n i ( t ) = x ( t ) y ( t ) ; t 2 [0 ; 1] ; (2) where b i s are the control points, and B n i ( t ) are the Bernstein polynomials of de gree n . F or a gi v en B ´ ezier curv e b ( t ) in (2), the of fset curv e b r ( t ) with of fset distance r 2 R + is gi v en by b r ( t ) = b ( t ) + r N ( t ) ; (3) where N ( t ) is the unit normal v ector of b ( t ) gi v en by N ( t ) = ( y 0 ( t ) ; x 0 ( t )) p ( x 0 ( t )) 2 + ( y 0 ( t )) 2 : (4) The error function e ( t ) is used to measure the error between N ( t ) and N a ( t ) and is gi v en by e ( t ) = ( y 0 ( t ) p x 0 2 ( t ) + y 0 2 ( t ) y 0 a ( t ) p x 0 2 a ( t ) + y 0 2 a ( t ) ) 2 + ( x 0 ( t ) p x 0 2 ( t ) + y 0 2 ( t ) x 0 a ( t ) p x 0 2 a ( t ) + y 0 2 a ( t ) ) 2 2. RESEARCH METHOD In this section, we present a ne w method of of fset curv e approximation of the n -th de gree B ´ ezier curv e by a curv e of de gree 3 . The best uniform approximation of the circular arc of de gree 3 of order 6 is presented in [7], see also [22-24]. The cubic approximation of circular arc p ( t ) has a parametrically defined polynom ial curv e gi v en by p ( t ) = 0 : 515647 + 5 : 99959 t 5 : 99959 t 2 0 : 874847 2 : 25031 t + 12 t 2 8 t 3 ; t 2 [0 ; 1] : (5) Let b ( t ) be a re gular planar B ´ ezier curv e of de gree n gi v en in (2) and N ( t ) be its unit normal v ector gi v en in (4) . As sho wn in Figure 1., gi v en an y B ´ ezier curv e b ( t ) then by the definition of the con v olutio, the tangent line of b ( t ) is parallel to the tangent line of N ( t ) which is the unit normal v ector for b ( t ) , 8 t 2 [0 ; 1] . Appr oximating of fset curves using B ´ ezier curves... (Abedallah Rababah) Evaluation Warning : The document was created with Spire.PDF for Python.
1650 r ISSN: 2088-8708 Figure 1. T angent of b ( t ) (thick) parallel to the tangent of N ( t ) (dashed) Thus b r N ( t ) = b ( t ) + r N ( t ) = b r ( t ) : Since N ( t ) is circular arc, the tangent line of N ( t ) is parallel to the tangent line of b ( t ) , then the approximation of N ( t ) is also circular arc and parallel to b ( t ) . Note that, N a ( s ( t )) and b ( t ) ha v e the same unit normal v ector . So, the of fset approximation is gi v en by b a r ( t ) = b r N a ( s ( t )) = b ( t ) + r N a ( s ( t )) ; t 2 [0 ; 1] : (6) The construction of the approximation N a ( s ( t )) , t 2 [0 ; 1] for the cubic B ´ ezeir curv e is considered. N a ( s ) = ( 2 : 25031 + 24 s 24 s 2 ) ; (5 : 99959 11 : 9992 s ) p (5 : 99959 11 : 9992 s ) 2 + ( 2 : 25031 + 24 s 24 s 2 ) 2 ; (7) where s = s ( t ) , t 2 [0 ; 1] is re gular reparametrization to mak e both curv es be gin and end at the same points. The curv e defined by b a r ( t ) = b N a ( s ( t )) = b ( t ) + r N a ( s ( t )) ; t 2 [0 ; 1] ; is the approximation of the of fset curv e by cubic B ´ ezier curv e where N a ( s ( t )) is as in (7) . The computation of the reparametrization, s = s ( t ) , where t 2 [0 ; 1] is considered. N a ( t ) and N ( t ) ha v e dif ferent parameters, both of them are circular arcs , b ut the y do not ha v e the same start and end points. Figure 2. sho ws N a ( t ) and N ( t ) for a B ´ ezier curv e b ( t ) . A reparametrization s = s ( t ) is pres ented, so that the curv e and its approximation be gin and end at the same points. Figure 2. N ( t ) (thick ) for B ´ ezier curv e and the approximation N a ( t ) (dashed) Int J Elec & Comp Eng, V ol. 10, No. 2, April 2020 : 1648 1654 Evaluation Warning : The document was created with Spire.PDF for Python.
Int J Elec & Comp Eng ISSN: 2088-8708 r 1651 Let b ( t ) be the curv e in (2) and N ( t ) be its unit normal v ector . The reparametrization is s ( t ) = t b + 1 t a ; t 2 [0 ; 1] ; where a; b are gi v en by the foll wing: the first step we find N (0) and N (1) for the gi v en curv e, b ( t ) . W e get to solv e the equations N a ( 1 a ) = x a ( 1 a ) y a ( 1 a ) = N (0) = x (0) y (0) (8) N a ( 1 b ) = x a ( 1 b ) y a ( 1 b ) = N (1) = x (1) y (1) : (9) By the symmetry of the approximation of the circular a rc, in equation (8) , x a ( 1 a ) and y a ( 1 a ) equal to zero at the same parameters, and (9) , x a ( 1 b ) and y a ( 1 b ) equal to one at the same parameters, then a equals the parameter in (8) and b equals the parameter in (9) . By solving the follo wing equations ( 2 : 25031 + 24( 1 a ) 24( 1 a ) 2 ) q (5 : 99959 11 : 9992( 1 a )) 2 + ( 2 : 25031 + 24( 1 a ) 24( 1 a ) 2 ) 2 = x (0) and (5 : 99959 11 : 9992( 1 a )) q (5 : 99959 11 : 9992( 1 a )) 2 + ( 2 : 25031 + 24( 1 a ) 24( 1 a ) 2 ) 2 = y (0) we get the v alue of the parameter a . And by solving ( 2 : 25031 + 24( 1 b ) 24( 1 b ) 2 ) q (5 : 99959 11 : 9992( 1 b )) 2 + ( 2 : 25031 + 24( 1 b ) 24( 1 b ) 2 ) 2 = x (1) and (5 : 99959 11 : 9992( 1 b )) q (5 : 99959 11 : 9992( 1 b )) 2 + ( 2 : 25031 + 24( 1 b ) 24( 1 b ) 2 ) 2 = y (1) we get the v alue of the parameter b . Then the approximation of N a ( t ) for the cubic case is N a ( t ) = ( 2 : 25031 + 24( t b + 1 t a ) 24( t b + 1 t a ) 2 ) ; (5 : 99959 11 : 9992( t b + 1 t a )) q (5 : 99959 11 : 9992( t b + 1 t a )) 2 + ( 2 : 25031 + 24( t b + 1 t a ) 24( t b + 1 t a ) 2 ) 2 : 3. RESUL TS AND AN AL YSIS The method is applied for the follo wing cubic parametric curv e: x ( t ) y ( t ) = 27 : 2688 t 3 + 341 : 56752 t 2 (1 t ) + 351 : 1 t (1 t ) 2 + 51 : 1(1 t ) 3 47 : 1461 t 3 + 338 : 8523975 t 2 (1 t ) + 333 : 324 t (1 t ) 2 + 21 : 4(1 t ) 3 ; (10) where t 2 [0 ; 1] and the unit normal v ector is gi v en by: N ( t ) = (35 : 772(1 t ) 2 +33 : 1704(1 t ) t +24 : 8811 t 2 ) p (35 : 772(1 t ) 2 +33 : 1704(1 t ) t +24 : 8811 t 2 ) 2 +(0 : (1 t ) 2 57 : 1949(1 t ) t 42 : 8962 t 2 ) 2 ( (0 : (1 t ) 2 57 : 1949(1 t ) t 42 : 8962 t 2 )) p (35 : 772(1 t ) 2 +33 : 1704(1 t ) t +24 : 8811 t 2 ) 2 +(0 : (1 t ) 2 57 : 1949(1 t ) t 42 : 8962 t 2 ) 2 : Appr oximating of fset curves using B ´ ezier curves... (Abedallah Rababah) Evaluation Warning : The document was created with Spire.PDF for Python.
1652 r ISSN: 2088-8708 Figure 3. represents the graph of the cubic parametric curv e and Figure 4. is the cubic para metric curv e with its of fset curv e computed by the formula. Figure 5. illustrates the parametric cubic curv e with the cubic approximation of the of fset curv e and the original of fset curv e. And Figure 6. illustrates the error between the of fset curv e and the approximation of the of fset curv e. Figure 3. The cubic parametric curv e Figure 4. Cubic parameteric curv e (thick) and its of fset curv e (dashed) By solving eqautions (8) and (9) , we get a = 2 ; b = 1 : 28862 : Figure 5. Cubic curv e (thick) and its of fset curv e (dashed) and the cubic approximation of the of fset curv e (dotted) Figure 6. Error between of fset curv e and the approximation of fset curv e 4. CONCLUSION In this article, cubic approximation of of fset curv e is established. The method is based on the best uniform approximation of the circular arc of de gree 3 with order 6 . The numerical e xamples re v eal ho w ef ficient this method is. The maximum er ror is 5 10 16 , thus the proposed method induced better outcomes than the e xisting methods. The results in this paper can be used to impro v e the results obtained in [25], see also the results in [26]. A CKNO WLEDGEMENTS The authors o we thanks for the re vie wers for in v aluable suggestions to impro v e an earlier v ersion of this paper . This research w as funded by Jordan Uni v ersity of Science and T echnology . Int J Elec & Comp Eng, V ol. 10, No. 2, April 2020 : 1648 1654 Evaluation Warning : The document was created with Spire.PDF for Python.
Int J Elec & Comp Eng ISSN: 2088-8708 r 1653 REFERENCES [1] Josef Hoschek. ”Spline approximation of of fset curv es. Computer Aided Geometric Design v ol. 5, Issue 1, pp. 33-40, 1988. [2] R. F arouki and T akis Sakkalis. ”Pythagorean hodographs. IBM Journal of Research and De v elopment v ol. 34, Issue 5, pp. 736-752, 1990. [3] Lee, In-Kw on, Myung-Soo Kim, and Gershon Elber . ”Planar curv e of fset based on circle approximation. Computer -Aided Design v ol. 28, Issue 8, pp. 617-630, 1996. [4] Ahn, Y oung Joon, Y eon Soo Kim, and Y oungsuk Shin. ”Approximation of circular arcs and of fset curv es by B ´ ezier curv es of high de gree. Journal of Computational and Applied Mathematics v ol. 167, Issue 2 , pp. 405-416, 2004. [5] Kim, Soo W on, Sung Chul Bae, and Y oung Joon Ahn. ”An algorithm for G2 of fset approximation based on circle approximation by G2 quadratic spline. Computer -Aided Design v ol. 73 , pp. 36-40, 2016. [6] Kim, Soo W on, Ryeong Lee, and Y oung Joon Ahn. ”A ne w method approximating of fset curv e by B ´ ezier curv e using parallel deri v ati v e curv es. Computational and Applied Mathematics v ol. 37, Issue 2, pp. 2053-2064, 2018. [7] A. Rababah, ”The best uniform cubic approximation of circular arcs with high accurac y . Communica- tions in Mathematics and Applications v ol. 7, Issue 1, pp. 37-46, 2016. [8] A. Rababah, The best uniform quadratic approximation of circular arcs with high accurac y , Open Mathe- matics 14 (1), 118-127, 2016. [9] A. Rababah, Quartic approximation of circular arcs us ing equioscillating error function, Int. J. Adv . Com- put. Sci. Appl 7 (7), 590-595, 2016. [10] T akashi Maeka w a, An o v ervie w of of fset curv es and surf aces, Computer -Aided Design, V olume 31, Issue 3, 1999, P ages 165-173. https://doi.or g/10.1016/S0010-4485(99)00013-5 [11] B. Pham, Of fset curv es and surf aces: a brief surv e y , Computer -Aided Design V olume 24, Issue 4, 1992, P ages 223-229. https://doi.or g/10.1016/0010-4485(92)90059-J [12] Huanxin Cao, Gang Hu, Guo W ei, and Suxia Zhang, Of fset Approximation of Hybrid Hyperbolic Poly- nomial Curv es, Results in Mathematics, 2017, V olume 72, Issue 3, 1055-1071. [13] Bhar g a v Bhatkalkar , Abhishek Joshi, Srikanth Prabhu, Sulatha Bhandary , Automated fundus image qual- ity assessment and se gmentation of optic disc using con v olutional neural netw orks. International Journal of Electrical and Computer Engineering (IJECE) V ol.10, No.1, 2020, pp. 816 827, ISSN: 2088-8708, DOI: 10.11591/ijece.v10i1.pp816-827. [14] A. Rababah, The best quintic Chebyshe v approximation of circular arcs of order ten, International Journal of Electrical and Computer En gi neering (IJECE)V ol. 9, No. 5, October 2019, pp. 3779-3785, ISSN: 2088- 8708, DOI: 10.11591/ijece.v9i5.pp3779-3785. [15] Rodiah, Sarifuddin Madenda, Diana T ri Susetianigtias, De wi Agushinta Rahayu, EtySutanty , Optic Disc and Macula Localization from Retinal Optical Coherence T omograph y and Fundus Image, International Journal of Electrical andComputer Engineering (IJECE) V ol.8, No.6, 2018, pp. 5050-5060, ISSN: 2088- 8708, DOI: 10.11591/ijece.v8i6.pp5050-5060. [16] G. Da w ay , Hana H. kareem, Ahmed Rafid Hashim, Pupil Detection Based on Color Dif ference and Circu- lar Hough T ransformHazim, International Journal of Electrical and Computer Engineering (IJECE) V ol.8, No.5, 2018, pp. 3278-3284, ISSN: 2088-8708, DOI: 10.11591/ijece.v8i5.pp3278-3284. [17] Gerald F arin, Curv es and surf aces for computer -aided geometric design: a practical guide. Else vier , 2014. [18] K. H ¨ ollig and J. H ¨ orner , ”Approximation and Modeling with B-Splines, SIAM, T itles in Applied Math- ematics , v ol. 132, 2013. [19] H. Prautzsch., et al., ”B ´ ezier and B-Spline T echniques, Springer 2002. [20] A. R ababah, L-2 de gree reduction of triangular B ´ ezier surf aces with common tangent planes at v er - tices. International Journal of Computational Geometry & Applications, V ol. 15, No. 5 (2005) 477-490. http://www .w orldscinet.com/ijcg a/15/1505/S02181959051505.html [21] A. Rababah,, Distances with rational triangular B ´ ezier surf aces. Applied Mathematics and Computation 160, ( 2005 ) 379-386. http://www .else vier .com/locate/amc [22] T or Dokk en, et al. ”Good approximation of circles by curv ature-continuous B ´ ezier curv es. Computer Aided Geometric Design v ol. 7, Issue 1-4, pp. 33-41, 1990. [23] Elber Gershon, In-Kw on Lee, and Myung-Soo Kim. ”Comparing of fset curv e approximation methods. IEEE computer graphics and applications v ol. 17, Issue 3, pp. 62-71, 1997. Appr oximating of fset curves using B ´ ezier curves... (Abedallah Rababah) Evaluation Warning : The document was created with Spire.PDF for Python.
1654 r ISSN: 2088-8708 [24] A. Rababah, ”T aylor theorem for planar curv es, Proceedings of the American Mathematical Society , v ol. 119, Issue 3, pp. 803–810, 1993. [25] C. Suw annapong and C. Khunboa, ”An Approximation Delay between Consecuti v e Requests for Con- gestion Control in Unicast CoAP-based Group Communication, International Journal of Electrical and Computer Engineering (IJECE) , v ol. 9, Issue 3, June 2019. DOI: http://doi.or g/10.11591/ijece.v9i3. [26] A. Rababah, ”Approximation v on K urv en mit Polynomen und Splines, Ph. Dissertation, Stuttg art Uni- v ersit ¨ at , W est German y , 1992. BIOGRAPHIES OF A UTHORS Abedallah Rababah is a professor of mathematics at United Arab Emirates Uni v ersity and is on lea v e from Jordan Uni v ersity of Science and T echnology . He is w orking in the field of Computer Aided Geome tric Design, abbre viated CA GD. In particular , his research is on de gree raising and reduction of B ´ ezier curv es and surf aces with geometric boundary conditions, Bernst ein polynomials, and their duality . He is kno wn for his research in describing approximation methods that significantly impro v e the standard rates obtained by classic al (local T aylor , Hermite) methods. He pro v ed the follo wing conjecture for a particular set of curv es of nonzero measure: Conjecture: A smooth re gular planar curv e can, in general, be approximated by a polynomial curv e of de gree n with order 2n. The method e xploited the freedom in the choice of the parametrization and achie v ed the order 4n/3, rather than n + 1. Generalizations were al so pro v ed for space curv es. Professor Rababah is also doing research in the fields of classical approximation theory , orthogonal polynomials, Jacobi-weighted orthogonal polynomials on triangular domains, and best unifor m approximati ons. Since 1992, He has been teaching at German, Jordanian, American, Canadian, and Emirates’ uni v ersities. He is acti v e in the editorial boards of man y journals in mathematics and computer science. Further info can be found on his homepage at ResearchGate: https://www .researchg ate.net/profile/Abedallah Rababah or at http://www .just.edu.jo/eportfolio/P ages/Def ault.aspx?email=rababah Moath J aradat graduated from J ordan Uni v ersity of Science and T echnology . He obtained both Bachelor and Ma ster De grees in Mathematics with research interests in the field of approximating of fset curv es using B ´ ezier curv es with high accurac y and numerical and approximations. Further info can be found on his homepage at ResearchGate: https://www .researchg ate.net/profile/Moad Jaradat Int J Elec & Comp Eng, V ol. 10, No. 2, April 2020 : 1648 1654 Evaluation Warning : The document was created with Spire.PDF for Python.