UNIT IV

IMAGE INTERPOLATION

Presented by
Mrs .S.Maria Seraphin Sujitha M.E,
Asst. Prof, ECE Dept,
St.Xavier’s Catholic College of
Engineering,
Chunkankadai.
UNIT IV
REGISTRATION AND IMAGE FUSION

Registration - Preprocessing, Feature selection -

points, lines, regions and templates Feature
correspondence - Point pattern matching, Line
matching, Region matching, Template matching.
Transformation functions - Similarity transformation
and Affine Transformation. Resampling – Nearest
Neighbour and Cubic Splines. Image Fusion -
Overview of image fusion, pixel fusion, wavelet based
fusion -region based fusion.
 RESAMPLING
 Digital images consist of a rectangular grid of
evently spaced pixels.
 Each pixel can be thought of as a measurement
or sample of the light from a subject.
 Resampling is the mathematical technique used
to create a new version of the image with a
different width and/or height in pixels.
 Increasing the size of an image is called
upsampling; reducing its size is called
downsampling.
.
 When images are upsampled, the number of pixels
increases, but, with reference to the original subject, new
image detail cannot be created that was not already
present in the original image.
 As a result, images normally become softer the more
they are enlarged since the amount of information per
pixel goes down.
 When images are downsampled, information in the
original image has to be discarded to make the image
smaller.
 If you downsample and then upsample an image, you will
not get all the original image detail back.
 Downsampling a soft image can make it appear sharper
even though it contains less information than the original.
 WHY IS RESAMPLING
IMPORTANT?

 Reducing the size of an image from a scanner or digital

camera for emailing or display on the web.
 Increasing the size of an image before or during the
printing process.
 HOW RESAMPLING WORKS
 Resampling a 2-dimensional image can be broken down into two
one-dimensional resampling passes.
1.In one pass, horizontal resampling is performed producing an
image with a different width but the same height.
2.In the next pass, this intermediate image is resampled vertically,
changing it height while leaving the width the same.
Advatages:
Computationally much more efficient than trying to combine the work
into a single pass.

 Upsampling involves interpolating between the existing pixels to

obtain an estimate of their values at the new pixel locations.
 Downsampling involves computing a weighted average of the
original pixels that overlap each new pixel.
 RESAMPLING METHODS

Interpolation is the process of using known data to

estimate values at unknown locations.

Most techniques work by

1) Computing new pixels as a weighted average of the
surrounding pixels. The weights depend on the distance
between the new pixel location and the neighboring pixels.
Advantages
a) simplest methods consider only the immediate neighbors.

2)More advanced methods examine more of the surround

pixels to attempt to produce a more accurate result.
 IMAGE INTERPOLATION

03/06/2024
Interpolation is the process of determining the values of a function
at positions lying between its samples.

8
CONT..

03/06/2024
 The ideal image interpolation algorithm should
preserve the qualitative characteristics of the output
image since interpolated images suffer from
artifacts, such as blurring, discontinuities in edges
and checkerboard effects.

9
 INTERPOLATION ALGORITHMS

03/06/2024
Common interpolation algorithms can be grouped into two
categories:
1.Adaptive
2.Non-adaptive.

Adaptive methods change depending on what they are interpolating,

Adaptive algorithms include many proprietary algorithms in licensed
software such as: Qimage, PhotoZoom Pro and Genuine Fractals.
Non-adaptive methods treat all pixels equally.
Non-adaptive algorithms include: nearest neighbor, bilinear, bicubic,
spline, sinc and others.
10
 TYPE OF INTERPOLATION

03/06/2024
3 main type of 2D interpolation
1. Nearest Neighbour Interpolation
2. Bilinear Interpolation
3. Cubic Splines Interpolation

11
 NEAREST NEIGHBOUR INTERPOLATION

03/06/2024
The method is called Nearest Neighbour
Interpolation because it assigns to each new
location the intensity of its nearest neighbour in the
original image new pixels as the value of the nearest
pixel.

Advantages
1.The nearest neighbor resampling method is very
simple.
Disadvantages
1.Blocky result when upsampling and a coarse, grainy 12
effect when downsampling.
 03/06/2024
The pictorial representation depicts that a 3x3 matrix is interpolated to 6x6
matrix. The values in the interpolated matrix are taken from the input matrix
(i.e) no new value is added.

13
CONT..

03/06/2024
Interpolation (Nearest
Input image
Neighbour)

14
BILINEAR INTERPOLATION

03/06/2024
 Bilinear interpolation use the four nearest neighbour
to assign the value.
 Bilinear interpolation, the assigned value is
obtained using the equation
 V(x,y)=ax+by+cxy+d

Where four coefficient are determined from the

four equation in four unknowns that can be written
using the four nearest neighbors of point(x,y).

15
03/06/2024
16
CONT..

03/06/2024
 Bilinear interpolation considers the closest 2x2 neighborhood
of known pixel values surrounding the unknown pixel. It then
takes a weighted average of these 4 pixels to arrive at its final
interpolated value. This results in much smoother looking
images than nearest neighbor.

 Interpolated value is simply their sum divided by

four.
17
03/06/2024
18
03/06/2024
19
CUBIC SPLINE INTERPOLATION

03/06/2024
 Using this process, a series of unique cubic
polynomials are fitted between each of the data
points, with the stipulation that the curve obtained be
continuous and appear smooth.
 Cubic splines can then be used to determine rates of
change and cumulative change over an interval.
 Cubic spline interpolation is based on the engineer’s
tool used to draw smooth curves through a number
of points.
 This spline consists of weights attached to a flat
surface at the points to be connected. A flexible strip
is then bent across each of these weights, resulting 20

in a pleasingly smooth curve.

CONT..

03/06/2024
The weights are the coefficients on the cubic
polynomials used to interpolate the data.
 These coefficients ’bend’ the line so that it passes
through each of the data points without any erratic
behavior or breaks in continuity.

21
03/06/2024
22
 PROCESS

03/06/2024
 The essential idea is to fit a piecewise function of
the form

23
CONT..

03/06/2024
The first and second derivatives of these n - 1 equations
are fundamental to this process,and they are

24
03/06/2024
25
03/06/2024
26
03/06/2024
27
03/06/2024
28
03/06/2024
29
03/06/2024
30
 plines are made up of the equations: –
 S 1 (x) = y 1 + b 1 (x - x 1 ) + c 1 (x - x 1 ) ^2 + d 1 (x - x

03/06/2024
1 ) ^3
 S 2 (x) = y 2 + b 2 (x - x 2 ) + c 2 (x - x 2 ) ^2 + d 2 (x - x
2 ) ^3 – … –
 S n-1 (x) = y n-1 + b n-1 (x - x n-1 ) + c n-1 (x – x n-1 ) ^2
+ dn-1(x – x n- 1 )^3
 With the following properties:
 S i (x i ) = y i and S i (x i+1 ) = y i+1 for i = 1,…,n - 1 –
Guarantees that the spline interpolates the data points S’
i-1 (x i ) = S’ i (x i ) for i = 2,…,n - 1 – Forces the slopes of
the parts of the spline to meet
 S’’ i-1 (x i ) = S’’ i (x i ) for i = 2,…,n - 1 – Forces the
curvature to agree
31
 03/06/2024
 So what we are looking to solve are the coefficients
of the cubic spline, a,b,c,d. To solve for these, we
use the formula: – S i (x) = y i + b i (x – x i ) + c i (x
– x i ) ^2 + d i (x – x i ) ^3 for n – 1 equations, where
n = number of points. The following finds the
coefficients for the equation: – a i = y i
 – δ i = x i+1 - x i

 – Δ i = y i+1 - y i –

 d i = (c i+1 – c i ) / 3δ i –

 b i = Δ i / δ i - δ i / 3 (2c i + c i+1 )

32
03/06/2024
33
 03/06/2024
Thank you

34