hw2_001

hw2_001 - W is a constant scaling factor equal to 100 R C...

This preview shows page 1. Sign up to view the full content.

ECE 178 HW #2 Due: Friday, Oct 9, 2009 Chapter 2 Problems 2.11-2.19, except 2.14 Programming assignment This is a MATLAB programming question. You need to familiarize yourself with the MATLAB environment first. It is strongly encouraged that you go through one of the many online MATLAB tutorials (see, for example, the book’s web site) and the exercises included in the class handout. Submit MATLAB code and results (you can use publish ) for the following: a) For years and years everyone has loved Lena Read in lena.gif (download from http://www.ece.ucsb.edu/~manj/ece178/lena.gif ) as a grayscale intensity image, S . Pay attention to the data type of S (uint8 vs. double) for consistency with your work in the following parts of the problem. b) But Lena only has eyes for MATLAB Add a special noise matrix, N , to the Lena image, S . The noise at each pixel is zero- mean Gaussian with variance as a function of its spatial position (x, y) as:
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: W is a constant scaling factor equal to 100 * R * C, the number of rows and columns in the input image S, respectively. Use of for or while loops is not permitted in computing the noise matrix N. Hints : Recall for a random variable G that var( α G) = α 2 var(G). Also r eview the functions meshgrid and randn for help in creating a variance matrix. c) Average SNR Compute the signal-to-noise ratio (a scalar value) averaged over all the pixels of the noised image O = S + N as: μ, the matrix of mean values of O is equal to S since the noise matrix N has zero means. Plot the SNR as you vary W from (10 * R * C) to (100 * R * C) over 10 equal points (see the plot and linspace functions .) Submit your plot along with a few examples of the noised image O for different values of W....
View Full Document

This note was uploaded on 12/28/2011 for the course ECE 178 taught by Professor Manjunath during the Fall '08 term at UCSB.

Ask a homework question - tutors are online