Lecture6-6 - 16827 - Line Scan Conversion

Chapter 3
Scan Conversion Algorithms

(Point and Line)
Four Major Tasks
• Four major task for rendering geometric objects
– modeling
– geometric processing Modeling
• transformations
• clipping Geometry
• shading
• hidden-surface removal
Rasterize
• projection
– Rasterization (Scan Conversion)
Display
– display
January 19, 2021 Computer Graphics 2

Scan Conversion
• Rasterization or Scan Conversion is defined as the
process of representing continuous graphic object as a
collection of discrete pixels.
– Various graphic objects are
• Point
• Line
• Rectangle, Square
• Circle, Ellipse
• Sector, Arc
• Polygons……
– Hardware, Software, Firmware Implementation

Scan Conversion
• Scan Conversion is required in order to convert vector
data to Raster format for a scan line display device
– convert each graphic object to a set of regular pixels.
– determine inside/outside areas when filling polygons.
– scan-convert curves

Scan Conversion
• The scan conversion process is interleaved with other
processes to deliver and improve the final image, some of
which are not entirely restricted to the discretization of vectors
but are involved in generally determining the colour of each
pixel in the raster. e.g:
– shading & illumination,
– hidden surface removal
– texture mapping
– depth testing

1. Scan Conversion of Point

2. Scan Conversion of Line
3. Scan Conversion of Circle
4. Scan Conversion of Ellipse
5. Scan Conversion of Polygons

Scan Converting a Point
A pixel (10, 6)
Y
X
A point (-12.3,
10.3, 0)
Z
Y
MODELLING CO-ORDINATES SCREEN COORDINATES

• Mathematically vectors are defined in an • a.k.a device co-ordinates, pixel co-ordinates
infinite, “real-number” Cartesian co- • On display hardware we deal with finite, discrete
ordinate system coordinates
• X, Y values in positive integers
• 0,0 is measured from top-left usually with +Y
pointing down
Scan Converting a Point
• Each pixel on graphic display does not represent a
mathematical point like P(2.6,3.33). But it can be
accommodated to the nearest position by applying few
mathematical functions such as
– CEIL P  (3,4)
– FLOOR P  (2,3)
– GREATEST INTEGER FUNCTION P  (3,3)
– ROUND P  (3,3)

1. Scan Conversion of Point

2. Scan Conversion of Line
3. Scan Conversion of Circle
4. Scan Conversion of Ellipse

Scan Converting a Line
How does a machine draw Lines?
1. Give it a start and end position.
2. Figure out which pixels to colour in between these…
– How do we do this?
– Line-Drawing Algorithms: DDA, Bresenham’s Algorithm

Scan Converting a Line
Line and its slope
– The slope of a line (m) is defined by its start and end coordinates
– The diagram below shows some examples of lines and their slopes
m=∞
m = -4 m=4
m = -2 m=2
m = -1 m=1
m = - 1 /2 m = 1/2
m = -1/3 m = 1/3

m=0 m=0
Line Drawing Algorithms
1. DDA Line Algorithm

2. Bresenham Line Algorithm

DDA Algorithm
1. Introduction
– Digital Differential Analyser is an
an algorithm for scan-converting lines
– The original differential analyzer
was a physical machine developed by
Vannevar Bush at MIT in the 1930’s
in order to solve ordinary differential equations.
– It calculates pixel positions along a line by taking unit step
increment along one direction and calculating corresponding
coordinate position based on the rate of change of the
coordinate (x or y ) (Incremental Approach)
DDA Algorithm
2. Basic Concept
– For each part of the line the following holds true:
y
m  y  mx
x
– If x = 1 i.e. 1 pixel then … y  m

– i.e. for each pixel we move right (along the x axis), we need to
move down (along the y-axis) by m pixels.
– In pixels, the gradient represents how many pixels we step
upwards (y) for every step to the right (x)

DDA Algorithm
3. Derivation
Assume that 0<m<1, x > 0 and y >0
For a point P(xi, yi ) on a line we know that
yi = mxi + b
At next position P(xi+1, yi+1 )
yi+1 = mxi+1+ b
Having unit step increment along x-axis means xi+1 = xi + 1
Therefore yi+1 = m(xi + 1)+ b
= mxi + m + b
= mxi + b + m
= yi + m
DDA Algorithm
4. Simple Algorithm
1. Input (x1,y1) and (x2,y2)
2. Let x = x1; y = y1;
m = (y2-y1)/(x2-x1);
3. Draw pixel (x, y)
4. WHILE (x < x2) //i.e. we reached the second
endpoint
5. {
x = x + 1; //step right by one pixel
y = y + m; //step down by m pixels
Draw pixel (ROUND(x), ROUND(y));
}

DDA ALGORITHM
5. Example
Sample at unit x:
xk 1  xk  x
 xk  1
Corresponding y pos.:
y k 1  yk  y
 y k  m  x
 y k  m  (1)

DDA ALGORITHM
5. Example
Sample at unit x:
xk 1  xk  x
 xk  1
y k 1  yk  y
 y k  m  x
 y k  m  (1)

DDA ALGORITHM
5. Example
Sample at unit x:
xk 1  xk  x
 xk  1
y k 1  yk  y
 y k  m  x
 y k  m  (1)

DDA ALGORITHM
5. Example
Sample at unit x:
xk 1  xk  x
 xk  1
y k 1  yk  y
 y k  m  x
 y k  m  (1)

DDA ALGORITHM
5. Example
Sample at unit x:
xk 1  xk  x
 xk  1
y k 1  yk  y
 y k  m  x
 y k  m  (1)

DDA ALGORITHM
5. Example
Sample at unit x:
xk 1  xk  x
 xk  1
y k 1  yk  y
 y k  m  x
 y k  m  (1)

DDA ALGORITHM
5. Example
Sample at unit x:
xk 1  xk  x
 xk  1
y k 1  yk  y
 y k  m  x
 y k  m  (1)
Consider endpoints:
P1(0,0), P2(7,4)

DDA ALGORITHM
6. Exercise
1. Consider endpoints:
P1(0,0), P2(6, 4)
Calculate the points that made up the line P1P2
2. Now, consider endpoints:
P3(0,0), P4(4, 6)
Calculate the points that made up the line P3P4
What happened with P3P4??????

DDA Algorithm
7. Y-Version (|m|>1)
– If the slope is very steep (i.e. y is

much greater than x) then there’s
a problem. i.e. m is much greater
than 1
– This is because we only draw one

pixel for each x value (each time
we step right)
– Solution: if the slope is too big, step

for y instead
DDA Algorithm
8. Generic solution
– Modify Basic idea y  mx m0
1
x  y
m
– Take unit step increment along direction having more rate of change, and
compute the coordinates for the other position to maximize the number of
points to be computed along the path of line and avoid discontinuities
xi  1 : xi  1
yi  1 : yi  m 0  m 1
or Why?
yi  1 : yi  1
1
xi  1 : xi  1 m
m
January 19, 2021 Computer Graphics discontinuity !! 26
DDA Algorithm - Generic solution
– What is m<0 ?
– We must see the sign of x and y
Various x>0 x>0 x<0 x<0
Cases y>0 y<0 y>0 y<0
xi+1 = xi + 1 xi+1 = xi + 1 xi+1 = xi - 1 xi+1 = xi - 1
|m|<1 yi+1 = yi + m yi+1 = yi - m yi+1 = yi + m yi+1 = yi – m
(or swap
points)
xi+1 = xi + 1/m xi+1 = xi + 1/m xi+1 = xi - 1/m xi+1 = xi - 1/m
|m|>1 yi+1 = yi + 1 yi+1 = yi - 1 yi+1 = yi + 1 yi+1 = yi – 1
(or swap
– Or can we have simple solution??????? points)
DDA Algorithm
1. Input (x1,y1) and (x2,y2)
2. Compute
dx = x2-x1 and dy = y2-y1
m = dx/dy
3. If abs(dx)>abs(dy)
step = abs(dx)
else
step = abs(dy)
4. Compute xinc = dx/step
yinc = dy/step
5. Initialize x = x1 and y = y1
6. Draw pixel (x, y)
7. For k = 1 to steps do //we reach the other endpoint
{
x = x + xinc;
y = y + yinc;
Draw pixel (ROUND(x), ROUND(y));
}
DDA Algorithm
8. Limitations
– Rounding integers takes time
– Variables y and m must be a real or fractional binary because

the slope is a fraction
– Real variables have limited precision, summing an inexact

slope m repetitively introduces a cumulative error buildup

DDA Algorithm
X Y Rounded { x, y }
Rounding Error
– Note that the actual pixel position is 1.0 0.33 { 1, 0 }
actually stored as a REAL number
2.0 0.66 { 2, 0 }
(in C/C++/java a float or a double)
3.0 0.99 { 3, 1 }
– But we Round Off to the nearest
4.0 1.32 { 4, 1 }
whole number just before we draw
the pixel.
– e.g. if m=.333 …

Line Drawing Algorithms
1. DDA Line Algorithm

2. Bresenham Line Algorithm

Bresenham’s Line Algorithm
1. Introduction
– One disadvantage of DDA is the ROUNDing part which can
be expensive
– Developed by Jack Bresenham
at IBM in the early 1960s
– One of the earliest algorithms in
computer graphics
– The Algorithm is based on essentially
the same principles but is completely
based on integer variables

2. Basic Concept
– Find the closest integer coordinates to the actual line path
using only integer arithmetic
– Candidate for the next pixel position
Specified
Line Path
0  m 1
Specified
Line Path
– No division, Efficient comparison, No floating point operations

3. Derivation
– The algorithm is derived for a line having slope 0< m < 1
in the first quadrant.
– Pixel position along the line are plotted by taking unit step
increments along the x-direction and determining y-
coordinate value of the nearest pixel to the line at each
step.
– Can be generalized for all the cases

– Let us assume that P(xk, yk ) is the currently plotted pixel.
Q(xk+1, yk+1 ) (xk+1, y ) is the next point along the actual path
of line. We need to decide next pixel to be plotted from two
candidate positions Q1(xk+1, yk ) or Q2(xk+1, yk+1)
Q2
yk  1 d2
yk  2 y Q
yk  1 Q2 d1
Q
y  mx  b yk Q1
yk P Q1 P
xk  1
xk xk  1 xk  2
0  m  1, xk  xl , k  l

Given the equation of line
y = mx + b
Thus actual value of y at x = xk+1 is given by
y = mxk+1 + b = m(xk + 1) + b
Let d1 = | QQ1|= distance of yk from actual value of y
= y – yk = m(xk + 1) + b – yk
d2 = | QQ2| = distance of actual value of y from yk +1
= yk+1 – y = (yk + 1)– [m(xk + 1) + b]
The difference between these 2 separations is
d1-d2 = 2m(xk + 1) + 2b –yk– (yk+ 1)
= 2m(xk + 1) – 2 yk + 2b – 1
we can define a decision parameter pk for the kth step to by
simplifying above equation such that the sign of Pk is the same as
the sign of d1-d2, but involves only integer calculations.
Define Pk = Δx ( d1-d2)
 x(2m( xk  1)  2 yk  2b  1)
y
 x(2 ( xk  1)  2 yk  2b  1)
x
 2y  xk  2x  yk  2y  x(2b  1)
 2 y  x k  2 x  y k  c
where c  2 Δy  Δx( 2 b  1 ) a constant

If pk< 0
 ( d1-d2) <0
 distance d1 is less than d2
 yk is closer to line-path
Hence Q1(xk+1, yk ) is the better choice
else
Q2(xk+1, yk+1) is the better choice
Thus if the parameter pk is negative lower pixel is plotted else

upper pixel is plotted
To put pk in the iterative form, we derived that
pk  2y.xk  2x. yk  c
Replacing k  k  1
pk  1  2y.xk  1  2x. yk  1  c
subtract pk from pk  1
pk  1  pk  2y ( xk  1  xk )  2x ( yk  1  yk )
pk  1  pk  2y  2x ( yk  1  yk )
 yk if pk  0
yk  1  
 yk  1 otherwise
 pk  2y if pk  0
 pk  1  
 pk  2y  2x otherwise
The first parameter p0 is directly computed as:
p0 = 2 Δy.x0 - 2 Δx.y0 + c
= 2 Δy.x0 - 2 Δx.y0 + 2 Δy + Δx (2b-1)
Since (x0,y0) satisfies the line equation , we also have

y0 = Δy/ Δx * x0 + b
b = y0 - Δy/ Δx * x0
Combining the above 2 equations , we will have

p0 = 2Δy – Δx
The constants 2Δy, 2Δy – Δx and 2Δy-2Δx are calculated once.
STEPS FOR BRESENHAM’S LINE DRAWING ALGORITHM (for |m| < 1.0)
1. Input the two line end-points (x0, y0) and (x1, y1)
2. Plot the point (x0, y0)
3. Compute Δx = x1 - x0 , Δy = y1 - y0
4. Initialize p0 = 2Δy – Δx
5. At each xk along the line, starting at k = 0, perform the
following test.
If pk < 0
the next point to plot is (xk+1, yk)
pk = pk +2Δy
else
the next point to plot is (xk+1, yk+1)
pk = pk +2Δy – 2Δx
6. Repeat step 5 (Δx – 1) times
7. Plot the point (x1, y1)
8. Exit
4. General solution
– The algorithm can be generalized for all slopes
Cases/ CaseI CaseII Case III Case IV
slope x>0 , y>0 x>0, y<0 x<0, y>0 x<0, y<0
p = 2dy-dx p = 2dy-dx swap points swap points
|m|<1 for x = x1 to x2 for x = x1 to x2
Case II Case I
if p<0 if p<0
p = p+ 2dy p = p+ 2dy
else { y=y+1 else { y=y-1
p = p+2(dy-dx)} p = p+2(dy-dx)}
p = 2dx-dy p = 2dx-dy swap points swap points
for y= y1 to y2 for y= y2 to y1
|m|>1 Case II Case I
if p<0 if p<0
p = p+ 2dx p = p+ 2dx
else { x=x+1 else { x=x+1
p = p+2(dx-dy)} p = p+2(dx-dy)}
5. Exercise
Calculate pixel positions that made up the line connecting endpoints:
(12, 10) and (17, 14).
1. (x0, y0) = ?
2. x = ?, y =?, 2y = ?, 2y – 2x =?
3. p0 = 2y – x =?
k pk (xk+1, yk+1)

5. Exercise
(12, 10) and (17, 14).
1. (x0, y0) = (12,10)
2. x = 5, y =4, 2y = 8, 2y – 2x =-2
3. p0 = 2y – x =3
k pk (xk+1, yk+1)
0 3

5. Exercise
(12, 10) and (17, 14).
1. (x0, y0) = (12,10)
2. x = 5, y =4, 2y = 8, 2y – 2x =-2
3. p0 = 2y – x =3
k pk (xk+1, yk+1)
0 3 (13, 11)
1 1 (14, 12)
2 -1 (15, 12)
3 7 (16, 13)
4 5 (17, 14)
5. Exercise: Trace for (20,10) to (30,18)
k pk (xk+1,yk+1)
18
17
16
15
14
13
12
11
10
20 21 22 23 24 25 26 27 28 29 30
Answer k pk (xk+1,yk+1)
18
0 6 (21,11)
17 1 2 (22,12)
16 2 -2 (23,12)
15 3 14 (24,13)
14 4 10 (25,14)
13 5 6 (26,15)
12 6 2 (27,16)
11 7 -2 (28,16)
10 8 14 (29,17)
20 21 22 23 24 25 26 27 28 29 30 9 10 (30,18)

Lecture6-6 - 16827 - Line Scan Conversion

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Lecture6-6 - 16827 - Line Scan Conversion

Uploaded by

Copyright:

Available Formats

Chapter 3

Scan Conversion Algorithms

January 19, 2021 Computer Graphics 2

January 19, 2021 Computer Graphics 3

January 19, 2021 Computer Graphics 4

January 19, 2021 Computer Graphics 5

1. Scan Conversion of Point

January 19, 2021 Computer Graphics 6

MODELLING CO-ORDINATES SCREEN COORDINATES

January 19, 2021 Computer Graphics 8

1. Scan Conversion of Point

January 19, 2021 Computer Graphics 9

January 19, 2021 Computer Graphics 10

January 19, 2021 Computer Graphics 11

1. DDA Line Algorithm

January 19, 2021 Computer Graphics 12

– If x = 1 i.e. 1 pixel then … y  m

January 19, 2021 Computer Graphics 14

January 19, 2021 Computer Graphics 16

January 19, 2021 Computer Graphics 17

January 19, 2021 Computer Graphics 18

January 19, 2021 Computer Graphics 19

January 19, 2021 Computer Graphics 20

January 19, 2021 Computer Graphics 21

January 19, 2021 Computer Graphics 22

January 19, 2021 Computer Graphics 23

What happened with P3P4??????

January 19, 2021 Computer Graphics 24

– If the slope is very steep (i.e. y is

– This is because we only draw one

– Solution: if the slope is too big, step

– Variables y and m must be a real or fractional binary because

– Real variables have limited precision, summing an inexact

January 19, 2021 Computer Graphics 29

January 19, 2021 Computer Graphics 30

1. DDA Line Algorithm

January 19, 2021 Computer Graphics 31

January 19, 2021 Computer Graphics 32

– No division, Efficient comparison, No floating point operations

January 19, 2021 Computer Graphics 34

January 19, 2021 Computer Graphics 35

January 19, 2021 Computer Graphics 37

Thus if the parameter pk is negative lower pixel is plotted else

Since (x0,y0) satisfies the line equation , we also have

Combining the above 2 equations , we will have

January 19, 2021 Computer Graphics 44

January 19, 2021 Computer Graphics 45

You might also like