M.Sc. (MATHEMATICS WITH APPLICATIONS IN COMPUTER SCIENCE) M.Sc. (MACS) O O
Term-End Examination December, 2011
MMTE-003 : PATTERN RECOGNITION AND IMAGE PROCESSING
Maximum Marks : 50
Time : 2 hours
Note : Attempt any five questions. All questions carry equal marks. Use of calculator is not allowed. 1.
(a) Propose a gray level slicing algorithm capable of producing the 4-th bit plane of an 8 - bit monochrome image. (b) Assume that the histogram of an image is
6
Gaussian (r m)2
pr(r)—
1 V2
Tr
cr
e 2 u2
Which transformation function would you use for histogram equalization ? 2.
(a) Briefly explain the following : (i)
Unsharp marking
(ii)
High boost filtering
MMTE-003
1
4
P.T.O.
(iii) High frequency emphasis filtering.
(b)
Give application of each of these filtering techniques. Given an image f(x,y) with Fourier transform 3 F(u,v) obtain the fourier transform of ( — 1)x+ Y. f(x ,y).
(c)
Given that the 2-D Fourier Transform is real and even, obtain the constraints on the image characteristics.
3.
(a) (i) In Image restoration, how are the noise 6 parameters estimated ? (ii) Assume that the noise is estimated as exponential, with mean 1.1. and variance o2 . How will you estimate the parameter 'a' of pdf of exponential noise ? (b) Explain the functioning of an adaptive, local 4 noise reduction filter.
4.
(a) Show that Sobel masks can be implemented by one pass of differencing mask of the form [ —101] (or its vertical counterpart) followed
5
by a smoothing mask of the form [1 2 1] (or its vertical counterpart). (b) (i) Explain the Hough transform for edge 5 linking. (ii) Why is the normal representation of line preferred ? Obtain the normal representation of the line y = — 2x + 1
MMTE-003
2
5. (a) Describe both spatial and frequency domain 4 enhancement techniques. (b) Describe briefly the watershed segmentation 3 Algorithm. (c) Define image segmentation. Discuss region 3 based segmentation in detail. 6. (a) The following pattern classes have Gaussion 5 probability density function : { (0, 0)T, (4, 0)T, (4, 4)T, (0, 4)T} and I (6, 6)T, (8, 6)T, (8, 8)T, (6, 0) T,nx Assume p(w1) = p(w2) = 1/2. Obtain the equation of Baye's decision boundary between these two classes. (b) Define Principal Component Analysis. 5 Derive the transformation matrix used for transformation where the data belongs to Rd. Interpret the transformation and its significance.
use for histogram equalization ? 2. (a) Briefly explain the following : (i) Unsharp marking. (ii) High boost filtering. 6. 4. MMTE-003 1 P.T.O.. Page 1 of 3 ...
why the filtering scheme is effective for the. applications it is used. 3. (a) Explain in detail the adaptive mean and 4. median filters. (b) Obtain mean and variance ...
resulting pattern becomes compact, which we call the compact image. We shall .... where fuy(u, v) is the p.d.f of the distorted image and C denotes the region we ...
Balau 1010. Balau 1011 ..... sion, and therefore the computation takes very long. However tests on the .... Distance (d) is fixed to 1 and the information extracted.
Processing of such huge image databases and retrieving the hidden patterns (or features) ... New data retrieval methods based on structural pattern recognition ...
J. Corso (SUNY at Buffalo) Introduction to Pattern Recognition 15 January 2013 4 / 41. Page 4 of 58. Pattern recognition Notes 1.pdf. Pattern recognition Notes ...
The Elements of Statistical Learning: Data Mining, Inference, and Prediction, ... Python for Data Analysis: Data Wrangling with Pandas, NumPy, and IPython.
the text including non-linear dimensionality reduction techniques, relevance feedback, semi- ... including real-life data sets in imaging, and audio recognition.
bAustralian National University, Canberra, ACT 0200, Australia ...... About the AuthorâHONGDONG LI obtained his Ph.D. degree from Zhejiang University, ...
The dramatic growth in practical applications for machine learning over the last ... The Elements of Statistical Learning: Data Mining, Inference, and Prediction, ...
The dramatic growth in practical applications for machine learning over the last ... The Elements of Statistical Learning: Data Mining, Inference, and Prediction, ...
