assignmnet i

1
Assignment-I Course Name: Pattern Recognition Batch: M.Tech (2 nd Semester) Year: 2012-2014 Due date: 25/02/2013 by 5 PM Q1. Consider an image in binary form. Binary form means the intensity values are only 0 (foreground) or 255 (background) for each pixel. Now develop an algorithm for converting a one-pixel thick 8-path to a 4-path. Extend the algorithm for converting a one-pixel thick m-path to a 4-path. Q2. Implement Hough transform to detect line segment exist among a set of points (pixels) in a two- dimensional Cartesian coordinate. The line will be represented as a normal form to reduce the problem of infinite slope when slope-intercept form is considered. Extend the algorithm for detecting a circle in a two-dimensional Cartesian coordinate. Q3. Consider a binary image. Show the resultant edge images if the following derivative operators are used. i. Ordinary ii. Roberts iii. 4-neighbour iv. Prewitt v. Sobel Compare the performance (based on some proposed evaluation parameters) of these gradient operators used in edge detection technique. Implementation and Evaluation Rules: 1. Use any language (C, C++, Matlab, etc.) for implementation. 2. Evaluation will be done based on the demonstration of the program code.

Upload: abhinav-choudhury

Post on 28-Apr-2015

74 views

Category:

Documents


3 download

DESCRIPTION

image processing and pattern recognition

TRANSCRIPT

Page 1: Assignmnet I

Assignment-I

Course Name: Pattern Recognition Batch: M.Tech (2nd

Semester)

Year: 2012-2014 Due date: 25/02/2013 by 5 PM

Q1.

Consider an image in binary form. Binary form means the intensity values are only 0 (foreground) or 255

(background) for each pixel. Now develop an algorithm for converting a one-pixel thick 8-path to a 4-path.

Extend the algorithm for converting a one-pixel thick m-path to a 4-path.

Q2.

Implement Hough transform to detect line segment exist among a set of points (pixels) in a two-

dimensional Cartesian coordinate. The line will be represented as a normal form to reduce the

problem of infinite slope when slope-intercept form is considered. Extend the algorithm for

detecting a circle in a two-dimensional Cartesian coordinate.

Q3.

Consider a binary image. Show the resultant edge images if the following derivative operators are used.

i. Ordinary

ii. Roberts

iii. 4-neighbour

iv. Prewitt

v. Sobel

Compare the performance (based on some proposed evaluation parameters) of these gradient operators used in

edge detection technique.

Implementation and Evaluation Rules:

1. Use any language (C, C++, Matlab, etc.) for implementation.

2. Evaluation will be done based on the demonstration of the program code.