SeatNo.

3216035
ITM (SLS) BARODA UNIVERSITY, VADODARA
SOCSET-B. Tecb-.Semester- VI (Regular) Examination-Summer, 2023
Course Code : C2620C2 · · Date : 26-05-2023
Course Name : Computer Vision Day : Thursday
Time : 10:00 am to 1:00 pm Total Marks : 100
Instructions:
1. All questions are mandatory. There are no external options.
2. Make suitable assumptions, wherever necessary, and state them clearly.
3. Use of Non-Programmable Calculator is allowed/Not allowed.
4. Figures to the right indicate maximum marks.

Q:1 Attempt the following Questions.

l. (20)
Computer vision is a discipline that studies how to reconstruct, interrupt and understand
a3d scene from its _ _ __
1
a) Id images b) 2 d images c) 3 d images d) 4 dimages
2. Low-level vision includes process image for feature extraction.
a) Yes b) No c) Can be yes or no d) Can not say 1
3. Which of the following is the local method for improving recall of an information
retrieval system?
a) Query expansion b) Relevance feedback
c) Ontology based model d) None of the above
4. Approaches to image processing that work directly on the pixels of incoming image
work in-:-:---:----
a) Spatial domain b) Inverse transformation 1
c) Transform domain d) None of the Mentioned
5. Which of the following operation is done on the pixels in sharpening the image, in the
spatial domain?
a) Differentiation b) Median 1.
c) Integration d) Average
6. Example of similarity approach in image segmentation is
a) edge based segmentation b) region based segmentation
c) both a and b d) None of the Mentioned 1

7. _ _ _ _ computes the output volume by computing dot product between all filters
and image patch.
a) _Input Layer b) Convolution Layer 1
c) Activation function Layer d) Pool Layer
8. Which of the following is NOT a constraint ofVR?
a) It can be costly
b) Resistance to new technology from teachers 1
c) Developing lessons/experiences can be time consuming
d) It can only be used for Science and Geography
9. Wha,tare the 3 types of Virtual Reality? .
'V,-31), non-immersive, digital b) Immersive, 3D, non-immersive
__,-c, ;rDigital, 1
. semi-immersive, projective ~ersive, semi-immersive, non-
unmers1ve
10, What are the names of the various colour image processing categories?
a) Pseudo-color and Multi-color processing
b) Half-color and pseudo-color processing 1
c) Full-color and pseudo-color processing
d) Half-color and full-color processing

Page 1 of4 •
. - ~-
 I I. In SVM, these functions take a lower dimensional input space and transform it to a
higher dimensional space,
a)Kemels b) Vector
c) Support Vector d) Hyperplane
12. A 4-input neuron has weights l, 2, 3 and 4. The transfer function is linear with the
constant of proportionality being equal to 2. The inputs are 4, 10, Sand 20 n:spectively,
What will be the ouii,ut?
a) 238 b)]6
c) 119 -.a(l23
13. Which of the following is true for neural networks?
(i) The training time depends on the size ofthe 'network.
(ii) Neural networks can be simulated on a conventional computer.
(iii) Artificial neurons are identical in operation to biological ones.

a) All of the mentioned b) (ii) is true

c) (i) and (ii) are true d) None of the mentioned
14. The Euclidean distance between two a set ofnwnerical attributes is called as?
a) Closeness b) Validation data
c) Error Rate d) None of these
15. In K Nearest NeighbolD' classification the Optimal value of 'k' training part considered
as? ,.
a) P b)P-1
c) 1-p d) I
16. The K-means algorithms
a) Requires the dimension if the feature space to be no bigger than the number of
samples
b) Has the smallest value of the objective function when k=l
c) Minimizes the within class variance for a given number of clusters
d) Converges to the global optimwn if and only if the ·initial means are chosen as
some of the samples themselves
17. Edge detection has fundamental.
a) 2 points b) 3 points
b) 4 points d) None of these
18. Hough transform is used for finding
a) Finding points in image c) Finding lines in image
b) Finding objects in image d) Finding moments of an image
19. Canny edge detection algorithm is based on
a) Ideal model c) Step edge
b) Real model d) Smoothing model
20. For point detection we use
a) First derivative c) Third derivative
b) Second derivative d) Both (a) and (b)

Q:2 Answer any Four out of Six Questions. (20)

• I. What is computer vision? Explain the of Diverse Computer Vision. 5
· 2. Explain the Virtual Reality and Augmented reality. 5
3. Explain Canny edge detection in detail. 5
, 4, Explain the OTSU Method in detail. • 5
S. Explain different applications of the clustering. How it is useful in patter recognition 5
explain in detail.
6. . Explain Artificial Neural network with example, 5

!'a~'! 2 cf4
Q:3 Answer Four out of Sb QuntloDJ.
What is the difference between low-level and high-level computer viJion? ,
(20)
~I. 5
,2. An 8x8 image is given in figure. Apply 3x3 min, max and median tilten on piul (5,5) 5 ..
with original value • 112. Give output value of pixel (S..,) for all tine fllten
senarately~ "\ "-
I 10 143 210 0 0 10 110 170
1.. 110 54 60 0 m 211 139 106
') 60 63 111 71 16~ 105 JO 70
'-I JO 11 255 o- 0 - IS 2SS 75
, 0-:- so 190 0 H2 0 6 200
0 130 JO 0 11 C
"' 75 200
60 0 150 10 200 61 255 so
255 105 Ill 41 211 113 201 106
• 3. Explam any two gradient based edge detectors. 5
, 4. What is image processing? Explain Global threshold method in detail. 5
5. Explain the Bayes theoiy with example and its advaotages and disadvantages. 5
6. 5
For a given k=2
- ,cuser
I t the oIIowing data set
Point x-a:w v-ui.s
I 7 6
2 2 6
3 3 8
4' 8 5
5 7 4
6 4 7
7 6 2
8 7 3
9 6 4
JO 3 4
Let choose that (3,4) and (7,4) !Ue the medoids. Suppose considering the
Manhanttan distance metric as the distance measure.

Q:4 Answer any Four out of Sb Questions. (20)

, I. Difference between Text based Image Retrieval and Content Based Image Retrieval 5
Also explain the types of features of content based image retrieval. .
- 2. Explain applications of Virtual Reality and Augmented reality, Justify with example 5
/ that this technology is will change the future.
3. Suppose the data mining task is to cluster points into three clusters, where the points are 5
Al (2, I 0),A2(2,5),A3(8,4),B I (5,8),B2(7,5),B3(6,4),CJ (1,2),C2(4,9) apply the K-means
clustering to solve the problem.
• 4. Explain Semi-Supervised Learning? Explain Examples and its disadvantages. 5
· 5. What is the image preprocessing? Explain the procedure of noise reduction, image 5
resizing.
,6. For a given image, show the results of rel!lon omwm2 · algorithm 5
I 0 7 8 7
0 I 8 9 8
0 0 7 9 8
0 I 8 8 9
I 2 8 8 9

Talce seed point as SI is 9 and S2 is I and threshold 'will be T<=4.

Page3of4 ·
~.,,