3d fractals fractals dominic rochon* universitÉ du quÉbec À trois-riviÈres may 2006 *research...
TRANSCRIPT
![Page 1: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/1.jpg)
3D FractalsDominic Rochon*
UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES
May 2006
*Research supported by CRSNG (Canada)
![Page 2: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/2.jpg)
Outline
I. The Birth of FractalsII. Bicomplex DynamicsIII. Bicomplex Distance Estimation
![Page 3: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/3.jpg)
The Birth of Fractals
Dominic Rochon*(Joint project with Josiane Lajoie)
UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES
May 2006
**Research supported by CRSNG (Canada)
![Page 4: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/4.jpg)
Outline : The Birth of Fractals
1. In the beginning2. The mathematical crisis3. About dimensions4. The birth of fractal geometry5. More about fractals
![Page 5: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/5.jpg)
IN THE BEGINNING
• First fractal images– Apollonius of Perga– Albretch Dürer
• Self similarity (Leibniz)
![Page 6: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/6.jpg)
The Apollonian Gasket(300 B.C.)
![Page 7: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/7.jpg)
Dürer Pentagon (1520)
![Page 8: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/8.jpg)
THE MATHEMATICAL CRISIS
• 1875 to 1925• ‟Pathological Monsters”• Begins with Karl Weierstrass
– Study of complex functions– A non-differentiable continuous curve!
![Page 9: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/9.jpg)
Helge von Koch
• Recall: Weierstrass (1875) showed an example of a non-differentiable continuouscurve.
• Helge von Koch (1904) suggested a simple construction of a continuous curvewithout a tangent.
![Page 10: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/10.jpg)
Koch Curve (1904)
![Page 11: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/11.jpg)
Waclaw Sierpinski
• Sierpinski gasket
• Generalization to all n sided polygons.
WaclawSierpinski
![Page 12: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/12.jpg)
Sierpinski square Sierpinski pentagon Sierpinski hexagon
Koch curve
Sierpinski Polygons
![Page 13: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/13.jpg)
3D Sierpinski
Sierpinski pyramid Menger – Sierpinski sponge
Source: http://fractals.nsu.ru/gallery_en.htm
![Page 14: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/14.jpg)
ABOUT DIMENSIONS
• First definition: Euclide (The Elements)• Problem: transformation from [0,1] to the
square.• Some definitions were suggested• Two types of dimension:
– Integer values (topologic)– Real values: Hausdorff (1919), Bouligand-
Minkowski (1929), Richarson (1960), Tricot (1982), …
![Page 15: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/15.jpg)
Box-counting Dimension
• Seems that it was developped by Hausdorff’sfollowers (around 1930)
• Idea: – Cover the figure with boxes of ε-length sides.– Determine the smallest number of boxes needed, in
function of ε noted N(ε).– Calculate the dimension using the following formula:
( )( )0
loglim
log 1/N
ε
εε→
⎛ ⎞⎜ ⎟⎜ ⎟⎝ ⎠
![Page 16: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/16.jpg)
Box-Counting Dimension of the Sierpinski Gasket
1
ε = 1
N(ε) = 1
ε = ½
N(ε) = 3
ε = ¼
N(ε) = 9
12
( ) 3d'où
log 3 log 3limlog 2 log 2
n
n
n
nn
N
d
ε
ε
→∞
=
=
⎛ ⎞= =⎜ ⎟
⎝ ⎠
![Page 17: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/17.jpg)
Coasts Length
• Richarson brought the problem of the coasts’ length. As an example:– Coast length between Spain and Portugal:
616 milles (Spain) or 758 milles (Portugal)• Coast length goes to infinity!
![Page 18: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/18.jpg)
Britain Coast Length
![Page 19: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/19.jpg)
THE BIRTH OF FRACTAL GEOMETRY
• So, we have:– Difficultly describable physics phenomenons;– ‟Pathological Monsters”.
• Mandelbrot’s idea: – Name those objects;– Study their communal properties.
• Creation of the word ‟fractal” (1975) whose rootmeans ‟broken” and ‟irregular”.
![Page 20: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/20.jpg)
Fractal Definition
• Different definitions depending of the authors.• Mandelbrot: ‟A fractal set (in a plane or in space)
is a set for which its Hausdorff-Besicovitchdimension is stricly greater then its topologicdimension.”
• Fractal caracteristics:– Its parts have approximately the same structure as
the whole;– Its form is extremly irregular or fragmented;– It contains new details on a large scaling range.
![Page 21: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/21.jpg)
Another Contribution• Mandelbrot noticed that fractals are everywhere in
nature:– Clouds are not spheres;– Mountains are not cones;– Islands are not circles; – …
• Natural elements description need an adaptedgeometry.
![Page 22: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/22.jpg)
MORE ABOUT FRACTALS
• There are different kinds of fractals.• Some ways to make fractals:
– Complex iterations– Iterated functions system (IFS)– L-systems
![Page 23: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/23.jpg)
Complex Iterations
• Julia (1918) suggested to iterate the complexpolynomial zn+1 = zn
2 + c for a constant number c.• The filled-Julia set associated to c (Kc) is formed
by the points z who generate a boundedsequence.
• The boundary of the filled-Julia set is the Julia set (Jc).
• The arrival of computers allowed us to visualizethose images.
![Page 24: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/24.jpg)
c = -0.8+0.168i
c = -0.344+0.64i
c = 0.328+0.048i
c = -1.096-0.264i
Filled-Julia Sets Associated to the Point cGaston
JuliaLouis Fatou
![Page 25: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/25.jpg)
Mandelbrot Set
• M = {c ∈ : Jc is connected}• We can generate the Mandelbrot set with
the same polynomial zn+1 = zn2 + c. This
time we fix z0 = 0 and we iterate for eachcomplex number c.
![Page 26: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/26.jpg)
Mandelbrot Set Benoit Mandelbrot
![Page 27: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/27.jpg)
Iterated Functions Systems (IFS)
• These iterations converge to a final image called ‟attractor” of the functions system.
• Theory is based on the Banach fixed point theorem. So,– The attractor exist;– The attractor is unique, no matter which
starting figure is chosen.
![Page 28: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/28.jpg)
Example of Fractal Construction Using IFS
![Page 29: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/29.jpg)
L-Systems
• Aristid Lindenmayer (1968) formallydescribed plant-growth.
• He used a character-string to describe a figure.
• Recursive process
![Page 30: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/30.jpg)
Example of a L-systemL-system: Plant
Axiom:Production rule:Parameter:
FF F[+F]F[-F]Fθ = 25,7o
Axiom:
FStep 1:
F [ + F ] F [ - F ] F
Step 2: …
(F[+F]F[-F]F) [+(F[+F]F[-F]F)](F[+F]F[-F]F)[-(F[+F]F[-F]F)](F[+F]F[-F]F)
Step 4:
…
![Page 31: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/31.jpg)
Chronological Summary
![Page 32: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/32.jpg)
Dominic Rochon*UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES
May 2006
*Research supported by CRSNG (Canada).
Bicomplex Dynamics
![Page 33: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/33.jpg)
Outline : Bicomplex Dynamics
Bicomplex NumbersBicomplex Dynamics
Bicomplex Mandelbrot SetBicomplex Filled-Julia SetsDefinition of the Tetrabrot
Generalized Fatou-Julia TheoremStrong Basin of Attraction of Cantor Sets in
Bicomplex Filled-Julia Sets in
∞4
3
![Page 34: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/34.jpg)
Quaternionic DynamicsIn 1982, A. Norton gave some straightforward algorithms for the generation and display in 3-D of fractal shapes. For the first time, iteration with quaternions (H) appeared. Subsequently, theoretical results have been treated for the quaternionicMandelbrot set defined by the quadratic polynomial in the quaternions of the form q2+c. However, S. Bedding and K. Briggs have established that there is “no interesting” dynamics for this approach and it does not play any fundamental role analogous to that for the map z2+c in the complex plane.
Recall: H:={ a + bi + cj + dk : i2 = j2 = k2=-1 }
![Page 35: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/35.jpg)
Quaternionic Dynamics
Author : Jean-François ColonnaSource : http://www.lactamme.polytechnique.fr
Author : Iñigo QuilezSource : http://rgba.scenesp.org
![Page 36: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/36.jpg)
Bicomplex NumbersIn 1892, in search for a development of special algebras, Corrado Segre(1860-1924) published a paper in which he treated an infinite set of algebras whose elements he called bicomplex numbers, tricomplex numbers, …, n-complex numbers.We define bicomplex numbers as follows:
t:={ a + bi1 + ci2 + dj : i12 = i2
2 = -1, j2 = 1 }
where i2j = ji2 = -i1, i1j = ji1 = -i2, i2i1 = i1i2 = j
and a, b, c, d ∈ .
![Page 37: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/37.jpg)
Bicomplex Numbers
We remark that we can write a bicomplex number a + bi1 + ci2 + dj as
(a + bi1) + (c + di1 ) i2 = z1 + z2i2
where ( ) { }21 2, : : 1 .z z x y∈ = + = −1 1 1i i i
![Page 38: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/38.jpg)
The idempotent basisIt is also important to know that every bicomplex number z1 + z2i2 has the following unique idempotent representation:
z1 + z2i2 = (z1 – z2i1)e1 + (z1 + z2i1)e2
where and .
This representation is very useful because: addition, multiplication and division can be done term-by-term. Also, an element will be non-invertible iff z1 – z2i1 = 0 or z1 + z2i1 = 0.
21 je1+
=2
1 je2−
=
![Page 39: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/39.jpg)
The idempotent basis
From the idempotent basis, it is now possible to define the notion of the bicomplex cartesian product.
Definition 1. We say that is a bicomplexcartesian set determined by X1 and X2 if
X = X1 xe X2 := { z1 + z2i2 ∈ T : z1 + z2i2 = w1e1 + w2e2, (w1, w2) ∈ X1 x X2 }.
X ⊆T
![Page 40: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/40.jpg)
Generalized Mandelbrot Set
Now, let us define a version of the Mandelbrot set for bicomplex numbers:Definition 2. Let Pc(w) = w2 + c where w, c ∈ Tand . Then the generalized Mandelbrot set for bicomplex numbers is defined as follows:
From this definition we obtain the following result:Theorem 1. The generalized Mandelbrot setM2 is connected.
( ) ( )( )( )wPPwP cn
cn
c1: −=
( ){ }2 : 0 .ncc P= ∈ → ∞M T
![Page 41: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/41.jpg)
Generalized Filled-Julia Sets
It is also possible to generalized the notion of filled-Julia sets for the bicomplex numbers:Definition 3. The generalized filled-Julia set for bicomplex numbers is defined as follows:
Finally, we obtain this relationship between the generalized Mandelbrot set and the generalized filled-Julia sets for the bicomplex numbers:Theorem 2. is connected.
( ){ }2, : nc cw P w= ∈ → ∞TK
2 2,cc∈ ⇔M K
![Page 42: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/42.jpg)
Display in 3-D
Previously, we established a version of the Mandelbrot set in dimension four. We are able now to give a version of the Mandelbrot set in dimension three using the definition for M2. The idea is to preserve the Mandelbrot set inside M2. Then, if we restrict the algorithm to the points of the form a+bi1+ci2 where a, b, c ∈ , we preserve the Mandelbrot set on two perpendicular complex planes and we stay in . This is the first argument to justify the following definition.
3
![Page 43: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/43.jpg)
Dynamics of Several Complex Variables
The polynomial Pc(w) = w2 + c is the following mapping of :
where w = z1 + z2i2 and c = c1 + c2i2. We note that this mapping is not a holomorphic automorphism of .
( )2 21 2 1 1 2 2, 2 .z z c z z c− + +
2
2
![Page 44: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/44.jpg)
• Definition 4. The “Tetrabrot” is defined as follows:
• It is possible to compute the infinite divergence layers of the Tetrabrot. We have to note at this step that each divergence layer will hide the others. For example, Fig. 2 is an illustration for the Tetrabrot of one of its divergence layers in correspondence with the divergence layer illustrated in Fig 1(A) for the Mandelbrot set.
( ){ }: 0 and 0 .nca b c d d P= + + + ∈ = → ∞1 2i i j TT
The Tetrabrot
![Page 45: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/45.jpg)
Fig. 1 Fig. 2
The Tetrabrot
![Page 46: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/46.jpg)
The Tetrabrot
• In fact, the Tetrabrot is inside Fig. 2. It is possible to see a part of the Tetrabrot (see Fig. 3) if we cut a piece of Fig. 2.
Fig. 2 Fig. 3
![Page 47: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/47.jpg)
The Tetrabrot
• In Fig 3, the colors represent the other divergence layers (see Figs 4, 5, 6 and 7). Figure 7 begins to be close to the set we wish to approach; then Fig 7 with its cut plane gives certainly a good idea of the Tetrabrot.
Fig. 3 Fig. 4 Fig. 5
Fig. 6
Fig. 7
![Page 48: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/48.jpg)
The Tetrabrot
Fig. 7 Fig. 8
• Moreover, we observe that the specific enlargement of Fig. 7 between A and B (Fig. 8) confirms that the Tetrabrot could be disconnected.
![Page 49: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/49.jpg)
The Tetrabrot• Finally, to define the Tetrabrot
we have put the last coordinate“j” equal to zero. In fact it is possible to do the same if we fix the last coordinate equal to a number different from zero. However, if we do that, we lose the beautiful symmetry of the Tetrabrot. Figure 9 gives an illustration of this phenomenon for a fixed “dj” with d ≠ 0.
Fig. 9
![Page 50: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/50.jpg)
The Cantor SetThe following classical theorem establishes a connection betweenthe Cantor sets and the filled-Julia sets:
Theorem 3 (P. Fatou and G. Julia) Let Kc be a filled-Julia set of the family of complex quadratic polynomials Pc(z) = z2 + c in the complex plane, and the basin of attraction of for Pc(z) = z2 + c.Then
(1) is connected;
(2) is a Cantor set.
0 c c∈ ⇔K K
0 c cA∈ ⇔ K
( ) /c cA ∞ = K
![Page 51: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/51.jpg)
The Cantor Set
Let us recall the definition of a Cantor set in :
Definition 5. A Cantor set is defined as a compact, perfect, totally disconnected subset in ..
Remark 1. Any such set is homeomorphic to the Cantor middle third set and therefore deserve the name of Cantor set.
n
.n
![Page 52: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/52.jpg)
Basin of attraction of ∞Now, let us define the concept of basin of attraction of ∞in the context of bicomplex numbers:
Definition 6. Let K2,c be a filled-Julia set of the family of bicomplex quadratic polynomials Pc(w) = w2 + c in T. We define A2,c(∞) = T\K2,c as the basin of attraction of ∞for Pc(w) = w2 + c. We note that
( ) ( ){ }2, .nc cA w P w∞ = ∈ →∞T
![Page 53: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/53.jpg)
• The next definition will be well justified in regard to the theorem below.
• Definition 7. We define
as the strong basin of attraction of ∞ for Pc(w) = w2 + cwhere c = (c1 – c2i1)e1 + (c1 + c2i1)e2. We note that:
( ) ( ) ( )( ) ( )
1 2 1 2
1 2 1 2
2,
c c
c c c e c c
c c e c c
SA
A A− −
− +
∞ = ×
= ∞ × ∞1 1
1 1
i i
i i
K K
( ) ( )2, 2, .c cSA A∞ ⊂ ∞
Generalized Fatou-Julia Theorem
![Page 54: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/54.jpg)
• The following theorem gives a characterizationof the filled-Julia sets for bicomplex numbersand introduces naturally the idea of Cantor sets in
• Theorem 4 (Fatou-Julia Theorem in t) Let K2,c be a filled-Julia set for bicomplex numbersand c ∈ t. Then(1) is connected;(2) is a Cantor set in T;(3) is disconnected
but not totally disconnected.
2, 2,0 c c∈ ⇔K K( )2, 2,0 c cSA∈ ∞ ⇔ K( ) ( )2, 2, 2,0 \c c cA SA∈ ∞ ∞ ⇔ K
4.
Generalized Fatou-Julia Theorem
![Page 55: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/55.jpg)
• Now it is possible to illustrate by figures in the connections between the various cases of the last theorem when the filled-Julia sets come from points around or inside the Tetrabrot. In fact, we are in the case 1 of the last theorem if and only if the filled-Julia sets come from points inside the Tetrabrot. The other cases are established from points inside the infinite divergence layers of the Tetrabrot. Fig 10 and 11 are an illustration of this phenomenonwhere the red zones are the points c which satisfy the case 3 and the other colors the points c which satify the case 2.
3
Generalized Fatou-Julia Theorem
![Page 56: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/56.jpg)
Fig. 10 Fig. 11
Generalized Fatou-Julia Theorem
![Page 57: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/57.jpg)
• More specifically, the Fig. 11 makes it possible to observe this phenomenon on the bottom inside the Tetrabrot for a specific divergence layer; the colors on the cut plane are an illustration of the other divergence layers of the Tetrabrot. We note also that the colors for the case 2 have been computed from the average of each divergence layer obtained from the number of iterations needed to know that zero is inside and( )∞− 1i21 ccA ( )
1 2.c cA + ∞
1i
Generalized Fatou-Julia Theorem
Fig. 11
![Page 58: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/58.jpg)
• The same process as for the Tetrabrot yields a version of the filled-Julia sets in . The next definition defined the bicomplex filled-Julia sets in .
• Definition 8. A bicomplex filled-Julia set in is defined as follows: ( c ∈ T )
( ){ }2, T: =0 and .nc cw a b c d d P w= = + + + ∈ → ∞1 2i i j L
The filled-Julia sets in .3
3
3
3
![Page 59: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/59.jpg)
The filled-Julia sets in .
• Figure 12 is an illustration of the filled-Julia set for the Tetrabrot at the same point c = 0.25 as the filled-Julia set D of Fig. 1. Hence, Fig. 12 is a kind of generalization of the filled-Julia set K0.25 in the complex plane.
Fig. 12Fig. 1
3
![Page 60: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/60.jpg)
Fig. 13
Fig. 14
Fig. 15
Fig. 16
• In the same manner, Figs. 13-16 are an illustration of the filled-Julia set at c = i1 for different divergence layers to infinity. We remark that Fig. 16 is a good approximation of this set and an interesting generalizationof Fig. 1(E).
Fig. 1
The filled-Julia sets in .3
![Page 61: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/61.jpg)
The same process can also be used when the filled-Julia sets are not connected. In that case we established these following results:
• Lemma 1. Let c ∈ T and K2,c a Cantor set in Then L2,c is compact and totally disconnected in
• Theorem 5. Let and Kc a Cantor set. Then, K2,c is a Cantor set in and Lc,2 is the union of a Cantor set in and a set which is at most countable.
{ }2( ) : : 1c x y∈ = + = −1 1i i i
3.
4
4
3
3“Cantor sets” in .3
![Page 62: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/62.jpg)
“Cantor sets” in .• Figures 17-34 give an illustration of such “Cantor sets” in . In fact,
Figs. 17-22 and 23-29 are respectively close to the filled-Julia sets of Fig. 12 and 13 without coming from points inside the Tetrabrot.
3
3
Fig. 12
Fig. 17 Fig. 18 Fig. 19
Fig. 20 Fig. 21 Fig. 22
![Page 63: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/63.jpg)
Fig. 23Fig. 22Fig. 21
Fig. 20Fig. 19Fig. 18
“Cantor sets” in .3
Fig. 13
• In each case, each step used a divergence layer to infinity closer to the true set. With this construction, we remark that we visually obtain a result similar to the geometric construction of the Cantor middle third set.
![Page 64: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/64.jpg)
“Cantor sets” in .3
Fig. 25 Fig. 26
Fig. 28 Fig. 29
Fig. 24
Fig. 27
• Morever, Fig. 29 is an ilustrationof L2,c for a filled-Julia set K2,c where:
• We note that this set in has connected com-ponents.
( ) ( )2, 2,0 \ .c cA SA∈ ∞ ∞
3
![Page 65: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/65.jpg)
Dominic Rochon*(Joint project with Étienne Martineau)
UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES
May 2006
*Research supported by CRSNG (Canada)
Bicomplex Distance Estimation
![Page 66: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/66.jpg)
Outline : Bicomplex Distance Estimation
• Bicomplex Numbers• Bicomplex Distance Estimation for the
Tetrabrot– Distance Formulas– Approximated Distance Formulas
• Applications– Ray-tracing– Exploration
![Page 67: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/67.jpg)
Bicomplex Numbers• The norm used on T is the Euclidean norm
(noted | |) of and the following formula is true over the set of bicomplex numbers:
12 2 2
1 2 1 21 2 .
2z z z z
z z⎛ ⎞− + +
+ = ⎜ ⎟⎜ ⎟⎝ ⎠
1 12
i ii
4
![Page 68: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/68.jpg)
Bicomplex Numbers• The set
will be called the set of hyperbolic numbers (also called duplex numbers) and
• Will be referred to as the modulus in j of w = z1 + z2i2.
{ }: | ,x y x y= + ∈jD
1 2 1 2: .w z z z z= − + + ∈1 1 1 2ji e i e D
![Page 69: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/69.jpg)
Bicomplex Numbers
• This specific modulus satisfies the following properties:
(1)
(2)
(3)
( )2Re ;w w w= =j j
1 2 1 2 1 2, .w w w w w w⋅ = ∀ ∈j j j
T
0 if and only if 0;w w= =j
![Page 70: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/70.jpg)
Distance Estimation for the Tetrabrot
• Let us begin with the following well known result about the distance estimation for the filled-Julia sets in the complex plane.
• Theorem 1. Let d(z, Kb) = inf{|z - a| : a ∈ Kb} be defined as the distance from z ∈ to the filled-Julia set Kb with b ∈ M. Then the distance d(z0, Kb) between z0 lying outside of Kb and Kb itself satisfies
Where G(z0) is the potential at the point z0.
( )( ) ( )
( ) ( )( )0
0 00
00
sinh 2sinh, .
'2 ' bG z
G z G zd z
G ze G z
⎡ ⎤ ⎡ ⎤⎣ ⎦ ⎣ ⎦< <K
![Page 71: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/71.jpg)
Distance Estimation for the Tetrabrot
• We will express the distance form a point w ∈ Tto a bicomplex filled-Julia set in terms of two distances in the complex plane (in i1).
• Lemma 1. Let d(w, K2,c) = inf{|w - a| : a ∈ K2,c}be defined as the “bicomplex” distance from w = z1 + z2i2 ∈ T to the bicomplex filled-Julia set K2,c where c = c1 + c2i2 ∈ T. Hence, d(w, K2,c) =
( ) ( )1
2
1 2 1 2
2 2
1 2 1 2, ,.
2c c c cd z z d z z− +
⎡ ⎤⎡ ⎤ ⎡ ⎤− + +⎣ ⎦ ⎣ ⎦⎢ ⎥⎢ ⎥⎢ ⎥⎣ ⎦
1 11 i 1 ii iK K
![Page 72: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/72.jpg)
Distance estimation for the Tetrabrot
• Definition 1. Let G1(z1 – z2i1) and G2(z1 + z2i1) be two electrostatic potentials. The bicomplex pot-ential, at a point is defined as
and
( )( ) ( )( )1 21 2 \ \b e bw z z= + ∈ ×2 1 1i i iK K
( ) ( ) ( )1 1 2 2 1 2:G w G z z G z z= − + + ∈1 1 1 2i e i e D
( ) ( ) ( )1 1 2 2 1 2' : ' ' .G w G z z G z z= − + + ∈1 1 1 2i e i e D
![Page 73: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/73.jpg)
Distance estimation for the Tetrabrot
• In T, the bicomplex logarithm ln(z1 + z2i1) is defined to be the inverse of the bicomplex exponential function:
With this definition of the bicomplex logarithm, it is possible to express the bicomplex potential in a similar way to that used for one complex variable. Let .
• Theorem 2. Let G : T\e K2,c →D be a bicomplex potential and c = (c1 – c2i1)e1 + (c1 + c2i1)e2 . Then
••
where is biholomorphic in terms of two complex variables.
( ) ( )1 2 12 2: cos sin .z z ze e z z+ = +⎡ ⎤⎣ ⎦
2i2i
( )( ) ( )( )1 2 1 22,\ : \ \e c c c e c c− += ×1 11 i 1 ii iK K KT
( ) ( )ln c jG w w wφ= ∀ ∈T
( ) ( )1 12,: \ \ 0,1 0,1c e c e eB Bφ → ×KT T
![Page 74: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/74.jpg)
Distance Formulas• We are now ready to state the major result of this talk.• Theorem 3. Let w0 = z1 + z2i1 ∈ T and c1 + c2i2 ∈ M2.
Then the distance d(w0, K2,c) between w0 lying outside of K2,c and K2,c itself satisfies:
(1) If w0 ∈ T\e K2,c ,
where G(w0) is the bicomplex potential at the point w0.
( )( ) ( )
( ) ( )( )0
0 00 2,
00
sinh 2sinh,
'2 ' cG w
G w G wd w
G we G w
⎡ ⎤ ⎡ ⎤⎣ ⎦ ⎣ ⎦< <K
![Page 75: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/75.jpg)
Distance Formulas(2) If
and
(3) If
-Similar to (2) -
( ) ( )( ) ( )1 1 2
1 1 20 2,
1 1 2
sinh,
2 2 'c G z z
G z zd w
e G z z−
−⎡ ⎤⎣ ⎦>−1
1i
1
i
iK
( ) ( )( )
1 1 20 2,
1 1 2
2 sinh, .
'c
G z zd w
G z z−⎡ ⎤⎣ ⎦<
−1
1
ii
K
( )( ) ( )1 2 1 20 \ ,c c e c cw − +∈ ×1 11 i ii K K
( )( ) ( )1 2 1 20 \ ,c c e c cw − +∈ ×1 11 i ii K K
![Page 76: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/76.jpg)
Approximated Distance Formulas
• Theorem 4. Let w0 = z1 + z2i2 ∈M2. Then, the distance d(w0, K2,c) between w0 lying outside of K2,c and K2,c itself approximatly satisfies:(1) If
where
and
0 2,\ ,e cw ∈ KT
( )0 2,12
ln, 2 ln
'2 'n
n n j nc n j
nnj
w w wd w www w
< <K
( )0: nn cw P w=
( )0
: .nn c
w w
dw P w ndw =
′ ⎡ ⎤= ∀ ∈⎣ ⎦
![Page 77: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/77.jpg)
Approximated Distance Formulas
(2) If
where
and
( )( ) ( )1 2 1 20 \ ,c c e c cw − +∈ ×1 11 i ii K K
( )( )
1, 2, 1, 2,0 2, 1
21, 2, 1, 2,
ln,
2 2 'n
n n n nc
n n n n
z z z zd w
z z z z
− −>
− −
1 1
1 1
i i
i iK
( )( )
1, 2,0 2, 1, 2,
1, 2,
2, ln
'n n
c n nn n n
z zd w z z
z z
−< −
−1
11
ii
iK
( ) ( )1, 2,
1, 2, ' : .n n
nn n c
z z z
dz z P zdz = −
⎡ ⎤− = ⎣ ⎦1
1i
i
( )1, 2, 1 2: nn n cz z P z z− = −1 1i i
![Page 78: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/78.jpg)
Approximated Distance Formulas
(3) If
- similar to (2) -
( )( ) ( )1 2 1 20 \ ,c c e c cw − +∈ ×1 11 i ii K K
![Page 79: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/79.jpg)
Ray-Tracing• Now we use the lower bound distance
estimation formula Dl in conjunction with ray-tracing to produce images of bicomplex fractals.
• Let be an unitary vector in and µ a point in T\K2,c. Now define
• By definition, no point in K2,c can be a member of such sequence.
v
{ } ( ), ,0
, ,, , , , 1 , , 1
:.
vv n
v n v n l v n
ZZ
Z Z D Z vµ
µµ µ µ
µ
− −
=⎧⎪= ⎨ = +⎪⎩
4
![Page 80: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/80.jpg)
Ray-Tracing• If we set the projection eye to µ and use as the
orientation of our ray, then
is our ray-tracing algorithm.
• Two things may happen, we may miss the fractal or we may converge i.e.
or
v
, ,lim v nnZµ→∞
( ), ,0
l v nn
D Zµ
∞
=
= ∞∑
( ) ( ), , , ,0
lim 0.l v n l v nnn
D Z D Zµ µ
∞
→∞=
< ∞⇒ =∑
![Page 81: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/81.jpg)
Exploration• The images of the fractals will be drawn on a
screen, noted S that is defined by four coplanar points in space. These points are our screen corner. We divide S into pixel according to the resolution desired for our image. The position of the eye µ, will be function of the position andsize of S. When we move S, µ will follow. We compute the first image of the object and while tracing the fractal, we keep stored the distance of the object.
![Page 82: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/82.jpg)
Exploration• To zoom into the region of interest, few steps
are necessary. First we must recenter the selected region by rotation from µ, then resize the screen to the region of interest. Next, using a fraction of distance to the object, we move S forward. A typical implementation could use the ratio screen size region of interest of the fractal distance. As we get closer to the fractal, we should lower the ε value to keep a good level of fractal details.
![Page 83: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/83.jpg)
More details available on my web site:
Dominic RochonDépartement de mathématiques et d’informatique
Université du Québec à Trois-RivièresC.P. 500 Trois-Rivières, Québec
Canada, G9A 5H7E-mail: [email protected]
www.3dfractals.com
![Page 84: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/84.jpg)
Tetrabrot inside the Tetrabrot
supplement
![Page 85: 3D Fractals Fractals Dominic Rochon* UNIVERSITÉ DU QUÉBEC À TROIS-RIVIÈRES May 2006 *Research supported by CRSNG (Canada) Outline I. The Birth of Fractals II. Bicomplex Dynamics](https://reader034.vdocument.in/reader034/viewer/2022051407/5b01066f7f8b9a952f8dc288/html5/thumbnails/85.jpg)
Tantrabrot
Kalachakra mandala Tetrabrot from the top
supplement