UNIT 2: 2D

Transformations,
Viewing &
Clipping
Introduction of
Transformations
 Computer Graphics provide the facility of viewing
object from different angles. The architect can study
building from different angles i.e.
 Front Evaluation
 Side elevation
 Top plan
 A Cartographer can change the size of charts and
topographical maps. So if graphics images are coded as
numbers, the numbers can be stored in memory.
 These numbers are modified by mathematical
operations called as Transformation.
 The purpose of using computers for drawing is to
provide facility to user to view the object from
different angles, enlarging or reducing the scale or
shape of object called as Transformation.
 Two essential aspects of transformation are given
below:
 Each transformation is a single entity. It can be denoted by
a unique name or symbol.
 It is possible to combine two transformations, after
connecting a single transformation is obtained, e.g., A is a
transformation for translation. The B transformation
performs scaling. The combination of two is C=AB. So C is
obtained by concatenation property.
 There are two complementary points of view for
describing object transformation.
 Geometric Transformation: The object itself is
transformed relative to the coordinate system or
background. The mathematical statement of this
viewpoint is defined by geometric transformations applied
to each point of the object.
 Coordinate Transformation: The object is held stationary
while the coordinate system is transformed relative to the
object. This effect is attained through the application of
coordinate transformations.
 Types of Transformations:
 Translation
 Scaling
 Rotating
 Reflection
 Shearing
1. Translation

 We perform a translation on a single coordinate point by

adding offsets to its coordinates so as to generate a new
coordinate position.

 It is the straight line movement of an object from one

position to another is called Translation. Here the
object is positioned from one coordinate location to
another.
 To translate a 2D position, we add translation distances tx and ty (that
pair is also called translation vector or shift vector) to the original
coordinates (x,y) to obtain the new coordinate position (x’,y’).
2. Scaling

 To alter the size of an object, we apply a scaling

transformation.
 A simple two-dimensional scaling operation is performed
by multiplying object position (x,y) by scaling factors sx
and sy to produce the transformed coordinates (x’,y’).
x’ = x . sx, y’ = y . sy
 Scaling factor sx scales an object in the x direction,
while sy scales in the y direction.
 The basic two-dimensional scaling equation can also be
written in the following matrix form:
 Any positive values can be assigned to the scaling
factors sx and sy.
 Values less than 1 reduce the size of objects;
values greater than 1 produce enlargements.
 Specifying a value of 1 for both sx and sy leaves
the size of objects unchanged.
 When sx and sy are assigned the same
value, a uniform scaling is produced,
which maintains relative object
proportions.
 Unequal values for sx and sy result in a
differential scaling that is often used in
design applications, where pictures are
constructed from a few basic shapes that
can be adjusted by scaling and positioning
transformation.
3. Rotation

 We generate a rotation transformation of an object by

specifying a rotation axis and a rotation angle.
 All the points of the object are then transformed to new
positions by rotating the points through the specified
angle about the rotation axis.
 A two-dimensional rotation of an object is obtained by
repositioning the object along a circular path in the xy
plane.
 Parameters for the two-dimensional rotation are the
rotation angle θ and a position (xr,yr), called the
rotation point or pivot point, about which the object is
to be rotated .
 The pivot point is the intersection position of the
rotation axis with the xy plane.
 A positive value for the angle θ defines a
counterclockwise rotation about the pivot point and
negative value rotates objects in the clockwise
direction.
 In the figure, r is the constant distance of the
point from the origin, angle φ is the original angle
position of the point from the horizontal, and θ is
the rotation angle.
 Using standard trigonometric identities we can
express as:
Homogeneous Coordinates
System
 Many graphics applications involve sequences of
geometric transformations.
 In design and picture construction applications, we
perform translations, rotations, and scalings to fit the
picture components into their proper positions.
 We can combine the multiplicative and translational terms for 2D
geometric transformations into a single matrix representation by
expanding the 2x2 matrix representations to 3x3 matrices.
 To express any 2D transformation as a matrix multiplication, we represent
each Cartesian coordinate position (x, y) with the homogeneous
coordinate triple (xh, yh, h), where:
 For 2D geometric transformations, we can choose the homogeneous
parameter h to be any nonzero value.
 Thus, there is an infinite number of equivalent homogeneous
representations for each coordinate point (x, y).
 A convenient choice is simple to set h = 1. Each 2D position is then
represented with homogeneous coordinates (x, y, 1).
 The term homogeneous coordinates is used in mathematics to refer to the
effect of this representation in Cartesian equations.
Other transformations:
Reflection
 It is a transformation which produces a mirror image of
an object. The mirror image can be either about x-axis
or y-axis. The object is rotated by180°.
 When the reflection axis is a line in the xy plane, the
rotation path about this axis is in a plane perpendicular
to the xy plane.
 For reflection axes that are perpendicular to the xy
plane, the rotation path is in the xy plane.
 Reflection about the line y = 0, the x axis, is accomplished with the
transformation matrix:

 Reflection about the y axis flips x coordinates while keeping y coordinates

the same. The matrix for this transformation is:
4. Reflection…
5. Shearing

 A transformation that distorts the shape of an object

such that the transformed shape appears as if the
object were composed in the internal layers that had
been caused to slide over each other is called a shear.

 Two common shearing transformations are those that

shift coordinate x values and those that shift y values.
5. Shearing…
 An x-direction shear relative to the x axis is produced with the
transformation matrix:

 Which transforms coordinate position as:

 We can generated x-direction shears relative to other reference lines

with:
Problems: 2D Geometric
Transformations
1. Translate a polygon with coordinates A(2,5), B(7,10)
and C(10,2) by 3 units in x direction and 4 units in y
direction.
2. A point (4,3) is rotated counterclockwise by an angle
45 degree. Find the rotation matrix and the resultant
point.
3. Scale the polygon with coordinate A(2,5), B(7,10) and
C(10,2) by 2 units in x direction and 2 units in y
direction.
Problems: 2D Homogeneous
coordinate
 Give a 3x3 homogeneous coordinate transformation
matrix for each of the following translations:
 Shift the image to the right 3 units
 Shift the image up 2 units
 Move the image down 0.5 unit and right 1 unit
 Move the image down 0.66 unit and left 4 units
Inverse Transformations

 For translation, we obtain the inverse matrix by

negating the translation distances.
 Thus, if we have 2D translation distance tx and ty, the
inverse translation matrix:

 This produces a translation in the opposite direction,

and the product of a translation matrix and its inverse
produces the identity matrix.
 An inverse rotation is accomplished by replacing the rotation angle by its
negative.

 Negative values for rotation angles generate rotations in a clockwise

direction, so the identity matrix is produced when any rotation matrix is
multiplied by its inverse.
 Because only the sine function is affected by the change in sign of the
rotation angle, the inverse matrix can also be obtained by interchanging
rows and columns.
 Now, we form the inverse matrix for any scaling transformation by
replacing the scaling parameters with their reciprocals.
 For 2D scaling with parameters sx and sy applied to the coordinate origin,
the inverse transformation matrix is:

 The inverse matrix generates an opposite scaling transformation, so the

multiplication of any scaling matrix with its inverse produces the identity
matrix.
2D Composite
Transformations
 With the matrix representations, we can set up a matrix
for any sequence of transformations as a composite
transformations matrix by calculating the matrix
product of the individual transformations.
 Forming products of transformation matrices is often
referred to as a concatenation, or composition, of
matrices.
 Because a coordinate position is represented with a
homogeneous column matrix, we must pre-multiply the
column matrix by the matrices representing any
transformation sequence.
 Also, because many positions in a scene are typically transformed by the
same sequence, it is more efficient to first multiply the transformation
matrices to form a single composite matrix.

 Thus, if we want to apply two transformations to point position P, the

transformed location would be calculated as: P’ = M2 . M1 . P
=M.P
Composite 2D Translations

 If two successive translation vectors (t1x , t1y) and (t2x

, t2y) are applied to a two-dimensional coordinate
position P, the final transformed location P’ is calculated
as:

 We can verify this result by calculating the matrix

product for the two associative groupings.
 Also, the composite transformation matrix for this
sequence of translations is:

 Which demonstrates that two successive translations

are additive.
Composite 2D Rotations

 Two successive rotations applied to a point P produce

the transformed position:

 By multiplying the two rotation matrices, we can verify

that two successive rotations are additive:
 so that the final rotated coordinates of a point can be
calculated with the composite rotation matrix as:
Composite 2D Scalings

 Concatenating transformation matrices for two

successive scaling operations in two dimensions
produces the following composite scaling matrix:
 The resulting matrix in this case indicates that
successive scaling operations are multiplicative.

 That is, if we were to triple the size of an object twice

in succession, the final size would be nine times that of
the original.
General 2D Pivot-Point
Rotation
 When a graphics package provides only a rotate function
with respect to the coordinate origin, we can generate
a two-dimensional rotation about any other pivot point
(xr , yr ) by performing the following sequence of
translate-rotate-translate operations:
1. Translate the object so that the pivot-point position
is moved to the coordinate origin.
2. Rotate the object about the coordinate origin.
3. Translate the object so that the pivot point is
returned to its original position.
which can be expressed in the form:
General 2D Fixed-Point
Scaling
 A transformation sequence to produce a two-
dimensional scaling with respect to a selected fixed
position (x f , y f ), when we have a function that can
scale relative to the coordinate origin only. This
sequence is:
1. Translate the object so that the fixed point coincides
with the coordinate origin.
2. Scale the object with respect to the coordinate origin.
3. Use the inverse of the translation in step (1) to return
the object to its original position
 Concatenating the matrices for these three operations produces the
required scaling matrix:

 This transformation is generated automatically in systems that provide a

scale function that accepts coordinates for the fixed point.
General 2D Scaling
Directions
 Parameters sx and sy scale objects along the x and y
directions.
 We can scale an object in other directions by rotating
the object to align the desired scaling directions with
the coordinate axes before applying the scaling
transformation.
 Suppose we want to apply scaling factors with values
specified by parameters s1 and s2 in the directions.
 To accomplish the scaling without changing the
orientation of the object, we first perform a
rotation so that the directions for s1 and s2
coincide with the x and y axes, respectively.

 Then the scaling transformation S(s1, s2) is

applied, followed by an opposite rotation to
return points to their original orientations.
 The composite matrix resulting from the product of these three
transformations is
Problems

 Show how shear transformation may be expressed in

terms of rotation and scaling.
 Apply the shearing transformation to square with A(0,0),
B(1,0), C(1,1) and D(0,1) as given below:
 Shear parameter value of 0.5 relative to the line yref = -1
 Shear parameter value of 0.5 relative to the line xref = -1
2-Dimensional Viewing and
Clipping
A graphics package allows a user to specify which part of a defined picture
is to be displayed and where that part is to be placed on the display
device.
 Any convenient Cartesian coordinate system, referred to as the world-
coordinate reference frame, can be used to define the picture.
 For a 2D picture, a view is selected by specifying a region of the xy plane
that contains the total picture or any part of it.
 The picture parts within the selected areas are then mapped onto
specified areas of the device coordinates.
 When multiple view areas are selected, these areas can be placed in
separate display locations, or some areas could be inserted into other,
larger display areas.
1. The 2D Viewing Pipeline

 A section of a two-dimensional scene that is selected for

display is called a clipping window because all parts of
the scene outside the selected section are “clipped”
off.
 The only part of the scene that shows up on the screen
is what is inside the clipping window.
 Sometimes the clipping window is alluded to as the
world window or the viewing window.
 Graphics packages allow us also to control the placement within the
display window using another “window” called the viewport.
 Objects inside the clipping window are mapped to the viewport, and it is
the viewport that is then positioned within the display window.
 The clipping window selects what we want to see; the viewport indicates
where it is to be viewed on the output device.
 By changing the position of a viewport, we can view objects at different
positions on the display area of an output device.
 Multiple viewports can be used to display different sections of a scene at
different screen positions.
 Usually, clipping windows and viewports are rectangles in standard
position, with the rectangle edges parallel to the coordinate axes.
 The mapping of a two-dimensional, world-coordinate scene description to
device coordinates is called a two-dimensional viewing transformation.
Sometimes this transformation is simply referred to as the window-to-
viewport transformation or the windowing transformation.
 Once a world-coordinate scene has been constructed, we could set up a
separate two-dimensional, viewing-coordinate reference frame for
specifying the clipping window.
 But the clipping window is often just defined in world coordinates, so
viewing coordinates for two-dimensional applications are the same as
world coordinates.
 Clipping is usually performed in normalized coordinates.
 This allows us to reduce computations by first concatenating the various
transformation matrices.
 Clipping procedures are of fundamental importance in computer graphics.
2. The Clipping Window

 To achieve a particular viewing effect in an application

program, we could design our own clipping window with
any shape, size, and orientation we choose.
 The simplest window edges to clip against are straight
lines that are parallel to the coordinate axes.
 Therefore, graphics packages commonly allow only
rectangular clipping windows aligned with the x and y
axes.
 Rectangular clipping windows in standard position are
easily defined by giving the coordinates of two opposite
corners of each rectangle.
Viewing-Coordinate Clipping
Window
 A general approach to the two-dimensional viewing
transformation is to set up a viewing-coordinate system
within the world-coordinate frame.
 This viewing frame provides a reference for specifying a
rectangular clipping window with any selected
orientation and position.
 To obtain a view of the world-coordinate scene as
determined by the clipping window of, we just need to
transfer the scene description to viewing coordinates.
 We choose an origin for a two-
dimensional viewing-coordinate frame
at some world position P0 = (x0, y0),
and we can establish the orientation
using a world vector V that defines the

yview direction. Vector V is called

the two-dimensional view up vector.
By changing the position of
the viewport
 Once we have established the parameters that defines the viewing-
coordinate frame, we transform the scene description to the viewing
system.
 The first step in the transformation sequence is to translate the viewing
origin to the world origin.
 Next, we rotate the viewing system to align it with the world frame.
 Object positions in world coordinates are then converted to viewing
coordinates with the composite two-dimensional transformation matrix:
World-Coordinate Clipping
Window
 A routine for defining a standard, rectangular clipping
window in world coordinates is typically provided in a
graphics-programming library.
 We simply specify two world-coordinate positions, which
are then assigned to the two opposite
 Once the clipping window has been established, the
scene description is processed through the viewing
routines to the output device.
 If we want to obtain a rotated view of a two-
dimensional scene, as discussed in the
previous section, we perform exactly the
same steps as described there, but without
considering a viewing frame of reference.

 Thus, we simply rotate (and possibly

translate) objects to a desired position and
set up the clipping window—all in world
coordinates.
By varying the size of the
viewports
3. Normalization and
Viewport Transformations
 With some graphics packages, the normalization and
window-to-viewport transformations are combined into
one operation.
 In this case, the viewport coordinates are often given
in the range from 0 to 1 so that the viewport is
positioned within a unit square.
 After clipping, the unit square containing the viewport
is mapped to the output display device.
Mapping the Clipping Window into a
Normalized Viewport

 To illustrate the general procedures for the

normalization and viewport transformations, we first
consider a viewport defined with normalized coordinate
values between 0 and 1.
 Object descriptions are transferred to this normalized
space using a transformation that maintains the same
relative placement of a point in the viewport as it had
in the clipping window.
 If a coordinate position is at the center of the clipping
window, for instance, it would be mapped to the center
of the viewport.
 Position (xw, yw) in the clipping window is mapped to position (xv, yv) in
the associated viewport.
 To transform the world-coordinate point into the same relative position
within the viewport, we require that:
 and the translation factors are:

 Because we are simply mapping world-coordinate positions into a viewport

that is positioned near the world origin, we can also derive using any
transformation sequence that converts the rectangle for the clipping
window into the viewport rectangle.
Mapping the Clipping Window into a
Normalized Square

 Another approach to two-dimensional viewing is to

transform the clipping window into a normalized
square, clip in normalized coordinates, and then
transfer the scene description to a viewport specified in
screen coordinates.
 Making these substitutions in the expressions for tx , ty, sx , and sy, we
have:

 Similarly, after the clipping algorithms have been applied, the normalized
square with edge length equal to 2 is transformed into a specified
viewport.
 The last step in the viewing process is to position the viewport area in the
display window.
 Typically, the lower-left corner of the viewport is placed at a coordinate
position specified relative to the lower-left corner of the display window.
Problems

1. Show that 2D reflection through X axis followed by 2D

reflection through the line Y = -X is equivalent to a
pure rotation about the origin.
2. Prove that successive 2D rotations are additive, i.e.
R(θ1) . R(θ2) = R(θ1 + θ2)
3. Prove that 2D rotation and scaling commute if Sx = Sy
or θ = nπ for integral n and that otherwise they do
not.
4. Clipping Algorithms

 Generally, any procedure that eliminates those portions

of a picture that are either inside or outside a specified
region of space is referred to as a clipping algorithm or
simply clipping.
 Usually a clipping region is a rectangle in standard
position, although we could use any shape for a clipping
application.
 The most common application of clipping is in the
viewing pipeline, where clipping is applied to extract a
designated portion of a scene (either two-dimensional
or three-dimensional) for display on an output device.
 Clipping algorithms are applied in two-dimensional viewing procedures to
identify those parts of a picture that are within the clipping window.
 Everything outside the clipping window is then eliminated from the scene
description that is transferred to the output device for display.
 An efficient implementation of clipping in the viewing pipeline is to apply
the algorithms to the normalized boundaries of the clipping window.
 The clipped scene can then be transferred to screen coordinates for final
processing.
2D Clipping
We explore two-dimensional algorithms for the following:

 Point clipping

 Line clipping (straight-line segments)

 Fill-area clipping (polygons)

 Curve clipping

 Text clipping
I. Point clipping
 For a clipping rectangle in standard position, we save 2D point P = (x, y)
for display.
 The points are said to be interior to the clipping window if:

 If any of these four inequalities is not satisfied, the point is clipped (not
saved for display).
 The equal sign indicates that points on the window boundary are included
within the window.
II. Line Clipping

 A line-clipping algorithm processes each line in a scene

through a series of tests and intersection calculations to
determine whether the entire line or any part of it is to
be saved.
 The expensive part of a line-clipping procedure is in
calculating the intersection positions of a line with the
window edges.
 Therefore, a major goal for any line-clipping algorithm
is to minimize the intersection calculations.
 To do this, we can first perform tests to determine
whether a line segment is completely inside the clipping
window or completely outside.
 It is easy to determine whether a line is completely inside a clipping
window, but it is more difficult to identify all lines that are entirely
outside the window.
 If we are unable to identify a line as completely inside or completely
outside a clipping rectangle, we must then perform intersection
calculations to determine whether any part of the line crosses the window
interior.
 We test a line segment to determine if it is completely inside or outside a
selected clipping-window edge by applying the point-clipping tests of the
previous section.
 When both endpoints of a line segment are inside all four clipping
boundaries, such as the line from P1 to P2 as in figure, the line is
completely inside the clipping window and we save it.
2D Line Clipping…
 And when both endpoints of a line segment are outside any one of the
four boundaries (as with line P3 P4 in Figure 9), that line is completely
outside the window and it is eliminated from the scene description.

 But if both these tests fail, the line segment intersects at least one
clipping boundary and it may or may not cross into the interior of the
clipping window.
 One way to formulate the equation for a straight-line segment is to use
the following parametric representation, where the coordinate positions
(x0, y0) and (xend, yend) designate the two line endpoints:
x = x0 + u(xend − x0)
y = y0 + u(yend − y0) 0≤u≤1
 We can use this parametric representation to determine where a line
segment crosses each clipping-window edge by assigning the coordinate
value for that edge to either x or y and solving for parameter u.
 However, if the value of u is within the range from 0 to 1, part of the line
is inside that border.
 We can then process this inside portion of the line segment against the
other clipping boundaries until either we have clipped the entire line or
we find a section that is inside the window.
 Processing line segments in a scene using the simple clipping approach
described in the preceding paragraph is straightforward, but not very
efficient.
II. Line Clipping: Cohen-
Sutherland
 This is one of the earliest algorithms to be developed
for fast line clipping, and variations of this method are
widely used.
 Processing time is reduced in the Cohen-Sutherland
method by performing more tests before proceeding to
the intersection calculations.
 Initially, every line endpoint in a picture is assigned a
four-digit binary value, called a region code, and each
bit position is used to indicate whether the point is
inside or outside one of the clipping-window
boundaries.
 For this ordering, the rightmost position (bit
1) references the left clipping-window
boundary, and the leftmost position (bit 4)
references the top window boundary.

 A value of 1 (or true) in any bit position

indicates that the endpoint is outside that
window border. Similarly, a value of 0 (or
false) in any bit position indicates that the
endpoint is not outside (it is inside or on) the
corresponding window edge.
 Sometimes, a region code is referred to as an “out” code because a value
of 1 in any bit position indicates that the spatial point is outside the
corresponding clipping boundary.
 Each clipping-window edge divides two-dimensional space into an inside
half space and an outside half space.
 Thus, an endpoint that is below and to the left of the clipping window is
assigned the region code 0101, and the region-code value for any endpoint
inside the clipping window is 0000.
 Bit values in a region code are determined by comparing the coordinate
values (x, y) of an endpoint to the clipping boundaries.
 Bit 1 is set to 1 if x < xwmin, and the other three bit values are
determined similarly.
 Instead of using inequality testing, we can determine the values for a
region-code more efficiently using bit-processing operations and the
following two steps:
 (1) Calculate differences between endpoint coordinates and clipping
boundaries.
 (2) Use the resultant sign bit of each difference calculation to set the
corresponding value in the region code.
 Once we have established region codes for all line endpoints, we can
quickly determine which lines are completely inside the clip window and
which are completely outside.
 Any lines that are completely contained within the window edges have a
region code of 0000 for both endpoints, and we save these line segments.
 Any line that has a region-code value of 1 in the same bit position for each
endpoint is completely outside the clipping rectangle, and we eliminate
that line segment.
 Lines that cannot be identified as
being completely inside or completely
out-side a clipping window by the
region-code tests are next checked
for intersection with the window
border lines.
 Therefore, several intersection
calculations might be necessary to
clip a line segment, depending on the
order in which we process the clipping
boundaries.
 It is possible, when clipping a line segment using this approach, to
calculate an intersection position at all four clipping boundaries,
depending on how the line endpoints are processed and what ordering we
use for the boundaries.
 To determine a boundary intersection for a line segment, we can use the
slope-intercept form of the line equation.
 For a line with endpoint coordinates (x0, y0)and (xend, yend), the y
coordinate of the intersection point with a vertical clipping border line
can be obtained with the calculation:
 Similarly, if we are looking for the intersection
with a horizontal border, the x coordinate can
be calculated as:

 with y set either to ywmin or to ywmax.

Cohen-Sutherland Line
Clipping Algorithm
II. Line Clipping: Midpoint
Subdivision

Algorithm
The Cohen-Sutherland line clipping algorithm requires the calculation of
the intersection of the line with the window edge.
 These calculations can be avoided by repeatedly subdividing the line at its
midpoint.
 If line is completely visible it its drawn and if it is completely invisible it is
rejected.
 If line is partially visible then it is subdivided in two equal parts. The
visibility tests are then applied to each half and this subdivision process is
repeated until we get completely visible and invisible line segments.
Midpoint Subdivision
Algorithm:
II. Line Clipping: Cyrus-Beck
Line Clipping
 The Cyrus–Beck algorithm is a generalized line
clipping algorithm.

 It was designed to be more efficient than the Cohen-

Sutherland algorithm, which uses repetitive clipping.

 Cyrus Beck is a line clipping algorithm that is made for

convex polygons.

 It allows line clipping for non-rectangular windows as in

Cohen-Sutherland line clipping method.
 Parametric line equation P0P1:P(t) = P0 + t(P1 - P0)
 Let Ni be the outward normal edge Ei. Now pick any arbitrary point PEi on
edge Ei then the dot product Ni.[P[Math Processing Error]t – PEi]
determines whether the point P[Math Processing Error]t is “inside the clip
edge” or “outside” the clip edge or “on” the clip edge.
 The point P[Math Processing Error]t is inside if Ni.[P[Math Processing Error]t –
PEi] < 0
 The point P[Math Processing Error]t is outside if Ni.[P[Math Processing
Error]t – PEi] > 0
 The point P[Math Processing Error]t is on the edge if Ni.[P[Math Processing
Error]t – PEi] = 0
 Ni.[P[Math Processing Error]t – PEi] = 0
 Ni.[ P0 + t(P1 - P0) – PEi] = 0 [Math Processing Error]ReplacingP(t with P0 +
t(P1 - P0))
 Ni.[P0 – PEi] + Ni.t[P1 - P0] = 0
 Ni.[P0 – PEi] + Ni∙tD = 0 (substituting D for [P1 - P0])
 Ni.[P0 – PEi] = - Ni∙tD

 It is valid for the following conditions −

 Ni ≠ 0 [Math Processing Error]
 D ≠ 0 (P1 ≠ P0)
 Ni∙D ≠ 0 (P0P1 not parallel to Ei)
Cyrus-Beck Line Clipping
Algorithm:
II. Line Clipping: Liang-
Barsky Line Clipping
 Faster line-clipping algorithms have been developed
that do more line testing before proceeding to the
intersection calculations.
 One of the earliest efforts in this direction is an
algorithm developed by Cyrus and Beck, which is based
on analysis of the parametric line equations.
 Later, Liang and Barsky independently devised an even
faster form of the parametric line-clipping algorithm.
 For a line segment with endpoints (x0, y0) and (xend, yend), we can
describe the line with the parametric form.

 In the Liang-Barsky algorithm, the parametric line equations are combined

with the point-clipping conditions to obtain the inequalities:
 We can define inequalities as:
x1 + u∆x >= xwmin
x1 + u∆x <= xwmax
y1 + u∆y >= ywmin
y1 + u∆y <= ywmax

 Now we can expressed as:

u∆x >= xwmin - x1 Change
u∆x <= xwmax - x1 Ok
u∆y >= ywmin - y1 Change
u∆y <= ywmax - y1 Ok
 Left

Right

Bottom

Top

 Any line that is parallel to one of the clipping-window edges has pk = 0 for
the value of k corresponding to that boundary, where k = 1, 2, 3, and 4
correspond to the left, right, bottom, and top boundaries, respectively.
Liang-Barsky Line Clipping
Algorithm
Advantages:

 It is more efficient that Cohen-Sutherland algorithm,

since intersection calculations are reduced.
 It requires only one division to update parameters t1
and t2.
 Window intersections of the line are computed only
once.
II. Line Clipping: Nicholl-
Lee-Nicholl Line Clipping
 Theory Assignment
 III. Polygon Fill-area
clipping
 Graphics packages typically support only fill areas that
are polygons, and often only convex polygons.
 To clip a polygon fill area, we cannot apply a line-
clipping method to the individual polygon edges directly
because this approach would not work.
 What we require is a procedure that will output one or
more closed polylines for the boundaries of the clipped
fill area, so that polygons can be scan-converted to fill
the interiors with the assigned color or pattern.
 We can process a polygon fill area against the borders of a clipping
window using the same general approach as in line clipping.

 We need to maintain a fill area as an entity as it is processed through the

clipping stages.

 Thus, we can clip a polygon fill area by determining the new shape for the
polygon as each clipping-window edge is processed.

 The interior fill for the polygon would not be applied until the final
clipped border has been determined.
 We first tested a line segment to determine whether it could be
completely saved or completely clipped, that will be the same with a
polygon fill area by checking its coordinate extents.

 If the minimum and maximum coordinate values for the fill area are inside
all four clipping boundaries, the fill area is saved for further processing.

 If these coordinate extents are all outside any of the clipping-window

borders, we eliminate the polygon from the scene description.
 When we cannot identify a fill area as being completely inside or
completely outside the clipping window, we then need to locate the
polygon intersection positions with the clipping boundaries.
 One way to implement convex-polygon clipping is to create a new vertex
list at each clipping boundary, and then pass this new vertex list to the
next boundary clipper.
 The output of the final clipping stage is the vertex list for the clipped
polygon.
 For concave-polygon clipping, we would need to modify this basic
approach so that multiple vertex lists could be generated.
1. Polygon Clipping:
Sutherland-Hodgeman
 To send the polygon vertices through each clipping stage
so that a single clipped vertex can be immediately
passed to the next stage.
 This eliminates the need for an output set of vertices at
each clipping stage, and it allows the boundary-clipping
routines to be implemented in parallel.
 The final output is a list of vertices that describe the
edges of the clipped polygon fill area.
 The general strategy is to send the pair of endpoints for each successive
polygon line segment through the series of clippers (left, right, bottom,
and top).
 As soon as clipper completes the processing of one pair of vertices, the
clipped coordinate values, if any, for that edge are sent to the next
clipper.
 Then the first clipper processes the next pair of endpoints.
 There are four possible cases that need to be considered when processing
a polygon edge against one of the clipping boundaries.
I. One possibility is that the first
edge endpoint is outside the
clipping boundary and the second
endpoint is inside.
II. Or, both endpoints could be inside
this clipping boundary.
III. Another possibility is that the first
endpoint is inside the clipping
boundary and the second endpoint
is outside.
1. If the first input vertex is outside this clipping-window border and the
second vertex is inside, both the intersection point of the polygon
edge with the window border and the second vertex are sent to the
next clipper.

2. If both input vertices are inside this clipping-window border, only the
second vertex is sent to the next clipper.

3. If the first vertex is inside this clipping-window border and the second
vertex is outside, only the polygon edge-intersection position with the
clipping-window border is sent to the next clipper.

4. If both input vertices are outside this clipping-window border, no

vertices
are sent to the next clipper.
2. Polygon Clipping: Weiler-
Atherton
This algorithm provides a general polygon-clipping approach that can be
used to clip a fill area that is either a convex polygon or a concave
polygon.
 The method was developed as a means for identifying visible surfaces in
the three-dimensional scene.
 Also use this approach to clip any polygon fill area against a clipping
window with any polygon shape.
 This algorithm traces around the perimeter of the fill polygon searching
for the borders that enclose a clipped fill region.
 To find the edges for a clipped fill area, we follow a path (either
counterclockwise or clockwise) around the fill area that detours along a
clipping-window boundary whenever a polygon edge crosses to the outside
of that boundary.

 The direction of a detour at a clipping-window border is the same as the

processing direction for the polygon edges.

 We can determine whether the processing direction is counterclockwise or

clockwise from the ordering of the vertex list that defines a polygon fill
area.
 In most cases, the vertex list is specified in a counterclockwise
order as a means for defining the front face of the polygon.

 Thus, the cross-product of two successive edge vectors that form a

convex angle determines the direction for the normal vector,
which is in the direction from the back face to the front face of
the polygon.

 If we do not know the vertex ordering, we could calculate the

normal vector, or we can locate the interior of the fill area from
any reference position.

 Then, if we sequentially process the edges so that the polygon

interior is always on our left, we obtain a counterclockwise
traversal.

 Otherwise, with the interior to our right, we have a clockwise

For a counterclockwise traversal of the polygon fill-area vertices, we apply the
following Weiler-Atherton procedure:

 Process the edges of the polygon fill area in a counterclockwise order until
an inside-outside pair of vertices is encountered for one of the clipping
boundaries; that is, the first vertex of the polygon edge is inside the clip
region and the second vertex is outside the clip region.

 Follow the window boundaries in a counterclockwise direction from the

exit-intersection point to another intersection point with the polygon. If
this is a previously processed point, proceed to the next step. If this is a
new intersection point, continue processing polygon edges in a
counterclockwise order until a previously processed vertex is encountered.

 Form the vertex list for this section of the clipped fill area.

 Return to the exit-intersection point and continue processing the polygon

edges in a counterclockwise order.
 For a clockwise edge traversal, we would use a clockwise clipping-window
traversal.
 Starting from the vertex labeled 1, the next polygon vertex to process in a
counterclockwise order is labeled 2.
 Thus, this edge exits the clipping window at the top boundary.
 We calculate this intersection position (point 1’) and make a left turn
there to process the window borders in a counterclockwise direction.
 Proceeding along the top border of the clipping window, we do not
intersect a polygon edge before reaching the left window boundary.
 Therefore, we label this position as vertex 1’’ and follow the left boundary
to the intersection position 1’’’.
 We then follow this polygon edge in a counterclockwise direction, which
returns us to vertex 1.

 This completes a circuit of the window boundaries and identifies the

verted list {1, 1’, 1’’, 1’’’} as a clipped region of the original fill area.

 Processing of the polygon edges is then resumed at point 1’.

 The edge defined by points 2 and 3 crosses to the outside of the left
boundary, but points 2 and 2’ are above the top clipping-window border
and point 2’ and 3 are to the left of the clipping region.
 Also, the edge with endpoints 3 and 4 is outside the left clipping boundary,
but the next edge (from endpoint 4 to endpoint 5) re-enters the clipping
region and we pick up intersection point 4’.

 The edge with endpoints 5 and 6 exits the window at intersection position
5’, so we detour the left clipping boundary to obtain the closed vertex list
{4’, 5, 5’}.

 We resume the polygon edge processing at position 5’, which returns us to

the previously processed point 1’’’.

 At this point, all polygon vertices and edges have been processed, so the fill
area is completely clipped.
IV. Curve Clipping
 We can first test the coordinate extents of an
object against the clipping boundaries to
determine whether it is possible to accept or
reject the entire object trivially.
 If not, we could check for object symmetries
that we might be able to exploit in the initial
accept/reject tests.
 For example, circles have symmetries
between quadrants and octants.
 An intersection calculation involves substituting a

clipping-boundary position (xwmin,

xwmax, ywmin, or yw max) in the
nonlinear equation for the object boundary and
solving for the other coordinate value.
 Once all intersection positions have been
evaluated, the defining positions for the object
can be stored for later use by the scan-line fill
procedures.
 The circle radius and the endpoints of the
clipped arc can be used to fill the clipped region,
by invoking the circle algorithm to locate
positions along the arc between the intersection
endpoints.
 Similar procedures can be applied when clipping a curved object against a
general polygon clipping region.

 On the first pass, we could compare the bounding rectangle of the object
with the bounding rectangle of the clipping region.

 If this does not save or eliminate the entire object, we next solve the
simultaneous line-curve equations to determine the clipping intersection
points.
V. Text Clipping

 The simplest method for processing character strings

relative to the limits of a clipping window is to use the
all-or-none string-clipping strategy.
 If all of the string is inside the clipping window, we
display the entire string. Otherwise, the entire string is
eliminated.
 This procedure is implemented by examining the
coordinate extents of the text string.
 If the coordinate limits of this bounding rectangle are
not completely within the clipping window, the string is
rejected.
 An alternative is to use the all-or-none character-clipping strategy.
 Here we eliminate only those characters that are not completely inside the
clipping window.
 In this case, the coordinate extents of individual characters are compared to
the window boundaries.
 Any character that is not completely within the clipping-window boundary is
eliminated.
 A third approach to text clipping is to clip the components of individual
characters.
 This provides the most accurate display of clipped character strings, but it
requires the most processing.
 If an individual character overlaps a clipping window, we clip off only the
parts of the character that are outside the window.

 Outline character fonts defined with line segments are processed in this
way using a polygon-clipping algorithm.

 Characters defined with bit maps are clipped by comparing the relative
position of the individual pixels in the character grid patterns to the
borders of the clipping region.