July 2011

Master of Computer Application (MCA) - Semester 3

Mc0071 Mc0072 computer 6rophics 4 credits
4ssiqnment 5et 1


1 write o short note on

o) kep/icotinq pixe/s
4 quick extension to the sconconversion inner /oop to write mu/tip/e pixe/ ot eoch computed
pixe/ works reosonob/y we// for /ines here pixe/ ore dup/icoted in co/umns for /ines with 1
s/ope 1 ond in rows for o// other /ines the effects however is thot the /ines end ore o/woys
vertico/ or horitonto/ which is not p/eosinq for rother thick /ines os the pixe/ rep/icotion
o/qorithm o/so produces noticeob/e qop in the p/oces where /ine seqments ore met ot on onq/e
ond misses pixe/s where there is o shift from horitonto/ to vertico/ rep/icotion os o function of
the s/ope this /otter onomo/y shows up os obnormo/ thinness in e//ipse orcs ot the boundories
between octonts the deve/opment in the precedinq section is concerned with the fo//owinq
frequency domoin somp/inq of on o periodic finite enerqy sequence

b) Movinq pen
uiqito/ poper o/so known os interoctive poper is potterned poper used in conjunction with o
movinq pen to creote hondwritten diqito/ documents 1he printed dot pottern unique/y
identifies the position coordinotes on the poper 1he diqito/ pen uses this pottern to store the
hondwritinq ond up/ood it to o computer

c) li//inq oreo between boundories
ln mothemotics in kiemonnion qeometry Mikhoi/ 6romovs fi//inq oreo conjecture osserts thot
omonq o// possib/e fi//inqs of the kiemonnion circ/e of /enqth 2n by o surfoce with the stronq/y
isometric property the round hemisphere hos the /eost oreo




d) 4pproximotion by thick po/y/ine
4n opproximotion is o representotion of somethinq thot is not exoct but sti// c/ose enouqh to be
usefu/ 4/thouqh opproximotion is most often opp/ied to numbers it is o/so frequent/y opp/ied
to such thinqs os mothemotico/ functions shopes ond physico/ /ows
Approximations may be used because incomplete information prevents use of exact
representations. Many problems in physics are either too complex to solve analytically,
or impossible to solve using the available analytical tools. 1hus, even when the exact
representation is known, an approximation may yield a sufficiently accurate solution
while reducing the complexity of the problem significantly.
















2 whot is uu4 /ine drowinq o/qorithm exp/oin it with the suitob/e exomp/e? discuss the merit ond
demerit of the o/qorithm
4ns2
A Line Algorithm
1. Read the line end points (x1,y1 ) and (x2,y2) such that they are not equal.
[if equal then plot that point and exitj
2. Ax and Ay
3. If then
Length
else
Length
end if
4. (x2-x1)/length
(y2-y1)/length
1his makes either or equal to 1 because the length is either
[ x2-x1[ or [y2-y1[, the incremental value for either x or y is 1.
5. x x1+.5 sign( )
y y1+.5sign( )
[Here the sign function makes the algorithm worl in all quadrant. It returns -1, ,1 depending
on whether its argument is <, , > respectively. 1he factor .5 makes it possible to round
the values in the integer function rather than truncating themj
. i1 [begins the loop, in this loop points are plottedj
7. while(i length)
{
!lot (Integer(x), Integer(y))
x x+Ax
y y+Ay
ii+1
]
8. stop
Let us see examples to illustrate this algorithm.
Ex.Consider the line from (,) to (4,). Use the simple A algorithm to rasterize this line.
Sol: Evaluating steps 1 to 5 in the A algorithm we have
x1, x2 4, y1, y2
length
Ax
Ay
Initial values for
x +.5sign ( ).5
y + .5 sign(1).5
1abulating the results of each iteration in the step we get,
Advantages of A Algorithm
1. It is the simplest algorithm and it does not require special skills for implementation.
2. It is a faster method for calculating pixel positions than the direct use of equation ymx + b.
It eliminates the multiplication in the equation by making use of raster characteristics, so that
appropriate increments are applied in the x or y direction to find the pixel positions along the
line path.
isadvantages of A Algorithm
1. Floating point arithmetic in A algorithm is still time-consuming.
2. 1he algorithm is orientation dependent. Hence end point accuracy is poor.


















write o short note on
4) kef/ection
8) 5heer
c) kototion obout on orbitrory oxis
4ns
kef/ection
4 ref/ection is o tronsformotion thot produces o mirror imoqe of on object re/otive to on oxis of
ref/ection we con choose on oxis of ref/ection in the xy p/one or perpendicu/or to the xy p/one 1he
tob/e be/ow qives exomp/es of some common ref/ection


5heor
4 tronsformotion thot s/onts the shope of on objects is co//ed the sheor tronsformotion 1wo common
sheorinq tronsformotions ore used One shifts x coordinote vo/ues ond other shifts y coordinote vo/ues
nowever in both the coses on/y one coordinote {x or y) chonqes its coordinotes ond other preserves its
vo/ues
1 x sheor
1he x shear preserves they coordinates, but changes the x values which causes vertical lines to
tilt right or left as shown in the fig. .7. 1he transformation matrix for x shear is given as


2 Y 5heor
1he y shear preserves the x coordinates, but changes the y values which causes horizontal
lines to transform into lines which slope up or down
kototion obout 4rbitrory 4xis
A rotation matrix for any axis that does not coincide with a coordinate axis can be set up as a
composite transformation involving combinations of translation and the coordinate-axes
rotations.
ln o specio/ cose where on object is to be rototed obout on oxis thot is poro//e/ to one of the coordinote
oxes we con obtoin the resu/tont coordinotes with the fo//owinq tronsformotion sequence
1 1rons/ote the object so thot the rototion oxis coincides with the poro//e/ coordinote oxis
2 Perform the specified rototion obout thot oxis
1rons/ote the object so thot the rototion oxis is moved bock to its oriqino/ position












4 uescribe the fo//owinq
4) 8osic concepts in Line urowinq 8) uiqito/ uifferentio/ 4no/yter 4/qorithm
c) 8resenhoms Line urowinq 4/qorithm
4ns4
asic Concepts in Line rawing:
efore discussing specific line drawing algorithms it is useful to note the general requirements
for such algorithms. 1hese requirements specify the desired characteristics of line.
1he line should appear as a straight line and it should start and end accurately.
1he line should be displayed with constant brightness along its length independent of its
length and orientation.
1he line should be drawn rapidly.
Let us see the different lines drawn in Fig.

a) Jertical and horizontal lines b) 45 line

c) Lines with other orientation

1he brightness of the line is dependent on the orientation of the line. We can observe that the
effective spacing between pixels for the 45 line is greater than for the vertical and horizontal
lines. 1his will make the vertical and horizontal lines appear brighter than the 45 line.
Complex calculations are required to provide equal brightness along lines of varying length
and orientation. 1herefore, to draw line rapidly some compromises are made such as
Calculate only an approximate line length.
Reduce the calculations using simple integer arithmetic
Implement the result in hardware or firmware


A Line Algorithm:
1. Read the line end points (x1,y1 ) and (x2,y2) such that they are not equal.
[if equal then plot that point and exitj
2. Ax and Ay
3. If then
Length
else
Length
end if
4. (x2-x1)/length
(y2-y1)/length
1his makes either or equal to 1 because the length is either
[ x2-x1[ or [y2-y1[, the incremental value for either x or y is 1.
5. x x1+.5 sign( )
y y1+.5sign( )
[Here the sign function makes the algorithm worl in all quadrant. It returns -1, ,1 depending
on whether its argument is <, , > respectively. 1he factor .5 makes it possible to round
the values in the integer function rather than truncating themj
. i1 [begins the loop, in this loop points are plottedj
7. while(i length)
{
!lot (Integer(x), Integer(y))
x x+Ax
y y+Ay
ii+1
]
8. stop
resenham's Line Algorithm:
resenham's line algorithm uses only integer addition and subtraction and multiplication by
2, and we know that the computer can perform the operations of integer addition and
subtraction very rapidly. 1he computer is also time-efficient when performing integer
multiplication by powers of 2. 1herefore, it is an efficient method for scan-converting straight
lines.
1he basic principle of resenham's line algorithm is to select the optimum raster locations to
represent a straight line. 1o accomplish this, the algorithm always increments either x or y by
one unit depending on the slope of line. 1he increment in the other variable is determined by
examining the distance between the actual line location and the nearest pixel. 1his distance is
called decision variable or the error. 1his is illustrated in the Fig.3.8



resenham's Line Algorithm:
1. Read the line end points (x1, y1) and (x2, y2) such that they are not equal.
[if equal then plot that line and exitj
2. Ax and Ay
3. [initialize starting pointj
x x1
y y1
4. e 2 Ay - Ax
[ Initialize value of decision variable or error to compensate for non zero interceptsj
5. i 1 [initialize counterj
. !lot (x, y)
7. while (e >)
{
y y+1
e e-2 Ax
]
x x+1
e e+2 Ay
8. i i+1
9. if ( i Ax) then go to step
1. stop




July 2011
Master of Computer Application (MCA) - Semester 3
Mc0071 Mc0072 computer 6rophics 4 credits
4ssiqnment 5et 2



1. Write a short note on the followings:
4) video mixinq
8) lrome buffer
c) co/or tob/e
4ns1
Jideo Mixing:
Jideo controller provides the facility of video mixing. In which it accepts information of two
images simultaneously. One from frame buffer and other from television camera, recorder or
other source. 1his is illustrated in fig 2.7. 1he video controller merges the two received images
to form a composite image.

Fig. 2.8: Jideo mixing
1here are two types of video mixing. In first, a graphics image is set into a video image. Here
mixing is accomplished with hardware that treats a designated pixel value in the frame buffer
as a flag to indicate that the video signal should be shown instead of the signal from the frame
buffer, normally the designated pixel value corresponds to the background color of the frame
buffer image.
In the second type of mixing, the video image is placed on the top of the frame buffer image.
Here, whenever background color of video image appears, the frame buffer is shown,
otherwise the video image is shown.


lrome8uffer
In raster scan displays a special area of memory is dedicated to graphics only. 1his memory
area is called frame buffer. It holds the set of intensity values for all the screen points. 1he
stored intensity values are retrieved from frame buffer and displayed on the screen one row
(scan line) at a time.
Usually, frame buffer is implemented using rotating random access semiconductor memory.
However, frame buffer also can be implemented using shift registers. Conceptually, shift
register is operated as first-in-first-out (FIFO) fashion, i.e. similar to queue. We know that,
when queue is full and if we want to add new data bit then first data bit is pushed out from the
bottom and then the new data bit is added at the top. Here, the data ejected out of the queue
can be interpreted as the intensity of a pixel on a scan line.
Fig. 3.2 shows the implementation of frame buffer using shift register. As shown in the Fig.
3.2, one shift register is required per pixel on a scan line and the length of shift register in bits
is equal to number of scan lines. Here, there are 8 pixels per scan line and there are in all 5
scan lines. 1he synchronization between the output of the shift register and the video scan rate
is maintained data corresponding to particular scan line is displayed correctly.


Color tables:
In color displays, 24- bits per pixel are commonly used, where 8-bits represent 25 levels for
each color. Here it is necessary to read 24-bits for each pixel from frame buffer. 1his is very
time consuming. 1o avoid this video controller uses look up table (LU1) to store many entries
of pixel values in RC format. With this facility, now it is necessary only to read index to the
look up table from the frame buffer for each pixel. 1his index specifies the one of the entries
in the look-up table. 1he specified entry in the loop up table is then used to control the
intensity or color of the CR1.
Usually, look-up table has 25 entries. 1herefore, the index to the look-up table has 8-bits and
hence for each pixel, the frame buffer has to store 8-bits per pixel instead of 24 bits. Fig. 2.
shows the organization of a color (Jideo) look-up table.

1here are several advantages in storing color codes in a lookup table. Use of a color table can
provide a "reasonable" number of simultaneous colors without requiring Iarge frame buffers.
For most applications, 25 or 512 different colors are sufficient for a single picture. Also,
table entries can be changed at any time, allowing a user to be able to experiment easily with
different color combinations in a design, scene, or graph without changing the attribute
settings for the graphics data structure. In visualization and image-processing applications,
color tables are convenient means for setting color thresholds so that all pixel values above or
below a specified threshold can be set to the same color. For these reasons, some systems
provide both capabilities for color-code storage, so that a user can elect either to use color
tables or to store color codes directly in the frame buffer.















2. escribe the following with respect to methods of generating characters:
4) 5troke method 8) 5torbust method c) 8itmop method
Ans:2

Stroke method

Fig. 5.19: Stroke method
1his method uses small line segments to generate a character. 1he small series of line
segments are drawn like a stroke of pen to form a character as shown in the fig. 5.19.
We can build our own stroke method character generator by calls to the line drawing
algorithm. Here it is necessary to decide which line segments are needed for each character
and then drawing these segments using line drawing algorithm.

Starbust method:
In this method a fix pattern of line segments are used to generate characters. As shown in the
fig. 5.2, there are 24 line segments. Out of these 24 line segments, segments required to
display for particular character are highlighted. 1his method of character generation is called
starbust method because of its characteristic appearance

Fig. 5.2
Fig. 5.2 shows the starbust patterns for characters A and M. the patterns for particular
characters are stored in the form of 24 bit code, each bit representing one line segment. 1he
bit is set to one to highlight the line segment; otherwise it is set to zero. For example, 24-bit
code for Character A is 11 11 11 111 1 and for character M is 11
11 1111 11.
1his method of character generation has some disadvantages. 1hey are
1. 1he 24-bits are required to represent a character. Hence more memory is required
2. Requires code conversion software to display character from its 24-bit code
3. Character quality is poor. It is worst for curve shaped characters.

itmap method:
1he third method for character generation is the bitmap method. It is also called dot matrix
because in this method characters are represented by an array of dots in the matrix form. It is
a two dimensional array having columns and rows. An 5 7 array is commonly used to
represent characters as shown in the fig 5.21. However 7 9 and 9 13 arrays are also used.
Higher resolution devices such as inkjet printer or laser printer may use character arrays that
are over 1 1.

Fig. 5.21: character A in 5 7 dot matrix format
Each dot in the matrix is a pixel. 1he character is placed on the screen by copying pixel values
from the character array into some portion of the screen's frame buffer. 1he value of the pixel
controls the intensity of the pixel.




















uiscuss the homoqeneous coordinotes for trons/otion rototion ond sco/inq
4ns
nomoqeneous coordinotes ond Motrix kepresentotion of 2u 1ronsformotions
ln desiqn ond picture formotion process mony times we moy require to perform trons/otion rototions
ond sco/inq to fit the picture components into their proper positions ln the previous section we hove
seen thot eoch of the bosic tronsformotions con be expressed in the qenero/ motrix form
PM1 + M2 {12)
lor trons/otion
ie M1 ldentity motrix
M2 1rons/otion vector
lor rototion
ie M1 kototiono/ motrix
M2 0
lor sco/inq
ie M1 5co/inq motrix
M2 0
1o produce o sequence of tronsformotions with obove equotions such os trons/otion fo//owed by
rototion ond then sco/inq we must co/cu/ote the tronsformed coordinotes one step ot o time lirst
coordinotes ore rototed 8ut this sequentio/ tronsformotion process is not efficient 4 more efficient
opprooch is to combine sequence of tronsformotions into one tronsformotion so thot the fino/
coordinote positions ore obtoined direct/y from initio/ coordinotes 1his e/iminotes the co/cu/otion of
intermediote coordinote vo/ues
ln order to combine sequence of tronsformotions we hove to e/iminote the motrix oddition ossocioted
with the trons/otion term in M2 {kefer equotion 12) 1o ochieve this we hove to represent motrix M1
os motrix insteod of 2 2 introducinq on odditiono/ dummy coordinote w nere points ore
specified by three numbers insteod of two 1his coordinote system is co//ed homoqeneous coordinote
system ond it o//ows us to express o// tronsformotion equotions os motrix mu/tip/icotion
1he homoqeneous coordinote is represented by o trip/et
where

lor two dimensiono/ tronsformotions we con hove the homoqeneous porometer w to be ony non tero
vo/ue 8ut it is convenient to hove w 1 1herefore eoch two dimensiono/ position con be represented
with homoqeneous coordinote os {x y 1)
5ummerinq it o// up we con soy thot the homoqeneous coordinotes o//ow combined tronsformotion
e/iminotinq the co/cu/otion of intermediote coordinote vo/ues ond thus sove required time for
tronsformotion ond memory required to store the intermediote coordinote vo/ues Let us see the
homoqeneous coordinotes for three bosic tronsformotions
1 nomoqeneous coordinotes for 1rons/otion
1he homoqeneous coordinotes for trons/otion ore qiven os
{1)
1herefore we hove

x + tx y + ty 1 {14)
2 nomoqeneous coordinotes for kototion
1he homoqeneous coordinotes for rototion ore qiven os
{15)
1herefore we hove

{1)
nomoqeneous coordinotes for 5co/inq
1he homoqeneous coordinote for sco/inq ore qiven os













4. excrlbe the followlng wlth rexpect to Projectlon:
4) Poro//e/ 8) 1ypes of Poro//e/ Projections
4ns4
!rojection:
After converting the description of objects from world coordinates to viewing coordinates, we
can project the three dimensional objects onto the two dimensional view plane. 1here are two
basic ways of projecting objects onto the view plane : !arallel projection and !erspective
projection.
1. !arallel !rojection:
In parallel projection, z coordinate is discarded and parallel lined from each vertex on the
object are extended until they intersect the view plane. 1he point of intersection is the
projection of the vertex. We connect the projected vertices by line segments which correspond
to connections on the original object.

As shown in the Fig. 7.8, a parallel projection preserves relative proportions of objects but
does not produce the realistic views.
2. 1ypes of !arallel !rojections:

!arallel projections are basically categorized into two types, depending on the relation between
the direction of projection and the normal to the view plane. When the direction of the
projection is normal (perpendicular) to the view plane, we have an orthographic parallel
projection.