CE5-R3 Page 1 of 2 January, 2010

paradepetΤεχνίτη Νοημοσύνη και Ρομποτική

5 Νοε 2013 (πριν από 3 χρόνια και 11 μήνες)

108 εμφανίσεις

CE5-R3 Page 1 of 2 January, 2010


Time: 3 Hours Total Marks: 100

a) Explain how 2D Fourier Transform is used to enhance an image in frequency domain.
b) What do you understand by convolution? Explain briefly the properties of convolution.
c) Describe briefly the DCT based JPEG compression technique.
d) What are the characteristic of selective and non-selective filters? Give examples of non-selective
e) Write short note on Run-Length Encoding Compression.
f) Define Morphology? What is morphological image processing? What are the applications of
Image morphing?
g) Explain Optical flow with suitable example.

a) What do you understand by camera calibration? Explain briefly.
b) Explain difficulties in 3D vision using intensity images as input.
c) i) Develop a procedure for computing the median of an n*n neighborhood.
ii) Propose a technique for updating the median as the center of the neighborhood is moved
from pixel to pixel.

a) Write an algorithm for basic global image thresholding. List applications of thresholding.
b) There are multiple ways of image segmentation like edge detection, thresholding, region-based
etc. Explain Region growing based segmentation.
c) Define spatial and gray-level resolution of an image. What will be the impact of decreasing
gray-level resolution of an image? What will be the impact of increasing spatial resolution of an
image? How to increase spatial resolution of an image?

a) Image filtering can be done in spatial and frequency domain. Explain a general procedure of
filtering in frequency domain.
b) Explain the HSV colour model. Explain its advantages and disadvantages, if any. Compare it with
RGB and CMYK models.

Answer question 1 and any FOUR from questions 2 to 7.

2. Parts of the same question should be answered together and in the same

CE5-R3 Page 2 of 2 January, 2010

a) Explain morphological algorithm for ‘Boundary Extraction’.
b) Huffman Code for the given data is given along with probability in the table.

Symbol Probability Code
A 0.1 011
B 0.4 1
C 0.06 01010
D 0.1 0100
E 0.04 01011
F 0.3 00

Decode the given sequence 010100111100.
c) Define coding, spatial and temporal redundancy.

a) What do you understand by Histogram Equalization? Explain the histograms corresponding to
four basic image types: Dark, Bright, low contrast and high contrast
b) Write Canny edge detector algorithm.

a) Define computer vision? What are the main tasks involved in any computer vision application?
Explain any two tasks.
b) What are the potential applications for society of the study of biological or computational vision