64149-mt----pattern Recognition

NR

Code No: 64149/MT M.Tech. – I Semester Supplementary Examinations, September, 2008 PATTERN RECOGNITION (Image Processing) Time: 3hours

Max. Marks:60 Answer any FIVE questions All questions carry equal marks ---

1.

What are the design concepts and methodologies for automatic pattern recognition? Explain clearly.

2.

What are the various geometrical properties of linear decision functions? Explain about hyper plane properties and dichotomies.

3.a) b)

What are the properties of joint density function? A random variable X has the density function 1 f (x) = , -2<x<2 4 = 0 , elsewhere Obtain (i) P (x<1) (ii) P (|x|>1) (iii) P (2x +3 > 5)

4.

What is Baye’s theorem? Explain how Baye’s classifier is useful for normal patterns.

5.

What is the role of histogram and window in classification techniques? What are the various decision making techniques? Explain.

6.

What do you understand by clustering and partitioning? Explain about ward’s method.

7.

What are the different distance measures? Explain about them clearly.

8.

Explain how pattern recognition techniques are useful in facial recognition. &_&_&_&