You are on page 1of 5

DIP (EE6131) ASSIGNMENT-3 AUTUMN 2021

COLOUR IMAGE PROCESSING

1. Use MATLAB to solve:


 Take a standard colour image and extract its R, G and B components. Is it
possible to get back the same colour image using the extracted R, G and B
components? If yes, show how?
 Find its (colour image) histogram. What do you interpret from the histogram?
 Find the chromaticity diagram and write the interpretation.
 Convert the RGB image to HSI, CMY and CMYK model.
 Find the histogram equalization of colour image without using any built-in
MATLAB function.

2. Consider the following 1000*1000 RGB image, in which the squares are fully saturated
red, green and blue (maximum intensity). An HSI image is generated from this image.
 Describe the appearance of each HSI component image.
 The saturation component of the HSI image is smoothed using an averaging mask
of size 250*250. Describe the appearance of the result.
 Repeat the same for hue component.
3. Assume that the monitor and printer of an imaging system are imperfectly calibrated. An
image that looks balanced on the monitor appears green in print. Describe general
transformations that might correct this imbalance.

4. Three colours are defined as follows in terms of the CIE XYZ primaries:
[Q1] = 0.5[X] + 0.3[Y]
[Q2] = 0.3[X] + 0.5[Y]
[Q3] = 0.1[X] + 0.2[Y]+0.7[Z]
(i) Identify, using correct notation, the tristimulus values of the colour [Q3] with respect
to the XYZ primaries.
(ii) Compute and identify with correct notation the XYZ chromaticities of [Q1], [Q2] and
[Q3] and plot them on the XYZ chromaticity diagram.

5. Derive the CMY intensity transformation function si = kri + (1 – k), i = 1, 2, 3, for (C, M,
Y), from its RGB counterpart si = kri, i = 1, 2, 3, for (R, G, B).

6. Three colours are defined as follows in terms of the CIE XYZ primaries:

[Q1] =1.25[X]+1.0[Y]+ 0.25[Z ]

[Q2] = 0.2[X]+ 0.6[Y]+0.2 [Z ]

[Q3] =0.5[X]+0.5[Y ]+1.5[Z ]

Compute and identity with correct notation the XYZ chromaticities of [Q1], [Q2] and
[Q3] and plot them on an XYZ chromaticity diagram. Indicate clearly on the diagram the
chromaticities of all colours that can physically be synthesized using linear combinations
of [Q1], [Q2] and [Q3] with positive coefficients. Show the chromaticity coordinates of
the colour [P] = 0.5[Q1] + 0.4[Q2] + 0.3[Q3] on the diagram.

IMAGE SEGMENTATION

7. Consider the image segment,

Based on the histogram, segment the image into two regions.


8. Apply the split-and-merge technique to segment the image given below.
9. Let f denote a 100 × 100 pixel image containing alternating black and white blocks (like a
chessboard), each of size 10 × 10 pixels. Consider the range of gray values to be all
integers from 0 (representing pure black) to 9 (representing pure white). Describe the
image obtained after exact histogram equalization with the attribute vector for every pixel
f(x, y) being (f(x, y), x, y), where x is the row number and y is the column number, with
lexicographic ordering.

10. The Sobel operator computes the following quantity at each pixel location in an image
array G[m,n] = | Gx[m,n]| + | Gy[m,n]|. The position of A[m,n] is the column m and row
n of the array. Write the 3 x 3 array for each mask, Mx and My. What mathematical
operation on an image array is approximated by the Sobel Operator? Show how the Sobel
operator is related to the mathematical operation.

11. Write MATLAB code for region growing image segmentation?

12. Derive the filtered response at the underlined pixel in the following image filtered using
the homogeneous mask area filter.

13. A person wishes to apply the Laplacian-of-a-Gaussian edge operator to an image f(m,n)
of size 256 x 256. The size of the edge operator is 32 x 32, and the origin is at its Centre.
Describe the procedure to perform the operation in frequency domain.
14. Plot the histogram of the image shown in Fig. 1. Segment the image using Iterative
Global Thresholding algorithm. Obtain the final threshold and number of iterations by
taking initial threshold value as 120. Comment on the results. Plot the histogram of the
final segmented image.

IMAGE SAMPLING & RECONSTRUCTION

15.
16.

You might also like