Professional Documents
Culture Documents
Set No. 1
JAWAHARLAL NEHRU TECHNOLOGICAL UNIVERSITY HYDERABAD
IV B.Tech.II Sem. II Mid-Term Examinations, April 2009
IMAGE PROCESSING
Objective Exam
Name:____________________________ Hall Ticket No. _____________________
Answer All Questions. All Questions Carry Equal Marks. Time: 20 Min. Marks: 20.
I.
1.
If the value of any given pixel can be predicated from the value of its neighbors, it is called
[
]
(a) Interpixel redundancy
(b) Psycho visual redundancy
(c) Coding redundancy
(d) bandwidth redundancy
2.
3.
4.
(a)
(b)
(c)
(d)
Dilation is analogous to
(a) Convolution
(b) correlation(c) Inverse
(d) Decorrelation
5.
6.
(c) Texture
(d) Graphs
7.
The perception algorithm converges in a finite number of iterations if the classes under
consideration are
[
]
(a) Linearly separable
(b) nonlinearly separable
(c) Identical
(d) Intersecting each other
8.
9.
An arrangement of descriptors is a
(a) Object
(b) Image
-2-
Set No. 1
[
(c) Pattern
(d) Pixcel
II.
11.
The process to address the problem of reducing the amount of data required to represent an
image is called as ___________
12.
13.
_____ techniques based on matching represent each class by a prototype pattern vector.
14.
For detection of isolated points, the sum of mask coefficients must be equal to
15.
III.
16.
[ True / False]
17.
[ True / False]
18.
[ True / False]
19.
Bayes classifier that is widely used in the classification of remotely sensed imagery generated
by multi-spectral scanners is
[ True / False]
20.
-oOo-
____
transform
[ True / False]
1.
2.
Quantization is an ____________operation
(a) Identity
(b) Reversible
(c) Inverse
[
(d) Irreversible
3.
[
(d) wavelet coding
4.
5.
6.
7.
(b)
(c)
(d)
Opening is
(a) Erosion followed by dilation
(c) Erosion followed by erosion
Syntactic pattern recognition uses
(a) Regular grammar (b) Type1
(d) patterns
Dilation of A by B is given by
(a)
(d) Type 3
8.
_______ have productions only of the form A , with A in N and in the set [
(a) Context free
(b) Regular
(c) Context sensitive
(d) Type 0
9.
10.
Cont2
-2-
Set No. 2
II.
11.
12.
13.
14.
15.
III.
16.
[ True / False]
17.
[ True / False]
18.
The perception algorithm converges in a finite number of iterations if the classes under
consideration are nonlinearly separable.
[ True / False]
19.
[ True / False]
20.
[ True / False ]
-oOo-
and known
1.
Assigning fewer bits to the more probable gray levels than to less probable one is called
[
]
(a) Fixed length coding
(b) Variable length coding
(c) Lossy coding
(d) Transform coding
2.
For which of the following compression techniques the compression ratio is more maximum
[
]
(a) Lossless (b) run length (c) Lossy
(d) bit plane
3.
JPEG is based on
(a) DCT
(b) DFT
4.
5.
6.
(c) DWT
(d) DST
(c) AB =1
(d) A B =1
(d) opening
In multi layer feed forward neural network the number of neuron nodes in the
(a) Number of pattern classes
(c) Dimensionality of input patterns
first layer is
[
]
7.
In 3 layer feed forward neural network, the number of neuron nodes in the output layer is
[
]
(a) Number of pattern classes
(b) any arbitrary number
(c) Dimensionality of input patterns
(d) infinity
8.
(a)
(b)
(c)
(d)
Cont2
-2-
Set No. 3
9.
10.
II.
11.
The process to address the problem of reducing the amount of data required to represent an
image is called as ___________
12.
13.
The best Bayes classifier for normal patterns places a general _____________
decision surface between each pair of pattern classes.
14.
weight in
15.
transform.
III.
16.
The methods that allow an image to be compressed and decompressed without losing
information is called as lossy compression.
[ True / False ]
17.
[ True / False ]
18.
[ True / False ]
19.
[ True / False ]
20.
Region based methods are more sensitive to noise than boundary based methods.
[ True / False ]
-oOo-
compresses each
order
1.
2.
3.
4.
5.
6.
(c) Complementation
Erosion of A by B is given by
(d) Inverse
[
(d) Trigonometric
[
(d) Closing
[
(a)
(b)
(c)
(d)
]
]
]
7.
8.
If the probability of a pattern X comes from class Wi is P(Wi /X) and it incurs a loss Lij if the
pattern classifier decodes that X came from Wj, then the conditional average risk is given by
[
]
(a)
(b)
(c)
(d)
[
]
(b) dilation , Identity and intersection
(d) opening
Cont2
-2-
Set No. 4
[
(c) Robert
(d) Sobel
(c) Block
[
(d) Cross section
]
]
II.
11.
The process to address the problem of reducing the amount of data required to represent an
image is called as ___________
12.
13.
The function that maps the output of the summing junction into the final output is called
__________ function.
14.
15.
____ grammars have productions only of the form A aB or A a with A and B in N and a
in
III.
16.
Assigning fewer bits to the more probable gray levels than to less probable one is called
Variable length coding.
[ True / False ]
17.
18.
In multi layer feed forward neural network the number of neuron nodes in the first layer is
equal to number of pattern classes.
[ True / False ]
19.
20.
The laplacian response is large whenever a pixel and the neighbors have the same value.
[ True / False ]
-oOo-
[ True / False ]
[ True / False ]