Image Processing - Dec 2013

Computer Engineering (Semester 7)

TOTAL MARKS: 80
TOTAL TIME: 3 HOURS (1) Question 1 is compulsory.
(2) Attempt any three from the remaining questions.
(3) Assume data if required.
(4) Figures to the right indicate full marks.

Justify/Contradict the following statement (any four) :-

1 (a) (i) Unit step sequence is a powersignal(5 marks) 1 (a) (ii) If the energy of the signal is finite its power is zero(5 marks) 1 (a) (iii) Brightness discrimination is poor at low levels of illumination(5 marks) 1 (a) (iv) Enhacement process does not add any information to the image.(5 marks) 1 (a) (v) All image compression techniques are invertible.(5 marks) 2 (a) Write an expression for 2-D DFT. What is its relationship with one dimention DFT? How one-dimensional FFT algorithm can be used to compute two dimensional DFT of an digital image.(10 marks) 2 (b) Define signals and systems and also give any 4 classification of Discrete Time Signals with examples.(10 marks)

Compare and constrast between the following (any two -)

3 (a) (i) Spatial Domain Processing and Transform Domain Processing.(5 marks) 3 (a) (ii) Image Enhancement and Image Restoration.(5 marks) 3 (a) (iii) Lossless and Lossy Compression(5 marks) 3 (b) Find the DFT of the given image.
[ ?egin{bmatrix} 0 & 1 & 2 & 1 \ 1 & 2& 3 & 2 \ 2 & 3 & 4 & 3 \ 1 & 2 & 3 & 2 end{bmatrix}](5 marks) 3 (c) Find the circular convolution of two sequence -
x₁(n)={1,-1,2,-4} and x₂(n)={1,2}(5 marks) 4 (a) Find the universe Z-transform of -
[xleft(z ight)=frac{z^3-4z^2+5z}{left(z-1 ight)left(z-2 ight)left(z-3 ight)}]
(i) ROC=|z|>3
(ii) ROC=|z|<1
ROC=2<|z|<3
(10 marks) 4 (b) What are the different types of redundancies in images.(5 marks) 4 (c) Explain Fidelity Criteria(5 marks) 5 (a) Explain filtering in spatial Domain.(10 marks) 5 (b) Explain segmentation based on discontinuities.(5 marks) 5 (c) Explain Hough-Transform and its application in detection of shapes.(5 marks) 6 (a) Explain the salient feature of the following codes :-
(i) Huffman code
(ii) Lossy predictive coding
(iii) Tranform coding.(10 marks) 6 (b) Explain with suitable example region splitting and merging techniques for segmentation.(10 marks) 7 (a) Sampling and Quantization.(5 marks) 7 (b) Edge linking and Boundry detection via graph theoretic technique.(5 marks) 7 (c) Image Restoration Model.(5 marks) 7 (d) Trimmed Average filter.(5 marks) 7 (e) Homomorphic filtering.(5 marks)