CS 696 Applied Computer Vision Homework Assignment 1 solved




I. Using Matlab [60%]
The goal of this problem set is to become familiar with basic Matlab commands, practice manipulating vectors and matrices, and try out basic image display and plotting functions. If you
are unsure what a Matlab function does, check the reference manual (at the command line, type
‘help’ and then the command name).

1.1. Read over the provided Matlab introduction code and its comments in intro.m. Open an interactive session in Matlab and test the commands by typing them at theprompt. (You might skip this step if you are an expert on Mablab).
1.2. Describe (in words where appropriate) the result of each of the following Matlab
commands. Use the help command as needed, but try to determine the output without
entering the commands into Matlab. Do not submit a screenshot of the result of typing
these commands.
a. >> x = randperm(5);

b. >> a = [1:10];
>> b = a([1:3:end]);

c. >> f = [1501:2000];
>> g = find(f > 1850);
>> h = f(g);

d. >> x = 22.*ones(1,10);
>> y = sum(x);

e. >> a = [1:1000];
>> b = a([end:-1:1]);

3. Given a 100 x 100 uint8 matrix A representing a grayscale image, write a few lines of
code to do each of the following. Try to avoid using loops.
a. Sort all the intensities in A, put the result in a single 10,000-dimensional vector x,and plot the values in x.
b. Display a figure showing a histogram of A’s intensities with 32 bins.
c. Create and display a new binary image the same size as A, which is whitewherever the intensity in A is greater than a threshold t, and black everywhereelse.
d. Generate a new image (matrix), which is the same as A, but with A’s meanintensity value subtracted from each pixel. Set any negative values to 0.
e. Use rand to write a function that returns the roll of a six-sided die.
f. Let y be the vector: y = [1:6]. Use the reshape command to form a newmatrix z whose first column is [1, 2, 3]’, and whose second column is [4, 5, 6]’.
g. Use the min and find functions to set x to the single minimum value that occursin A, and set r to the row it occurs in and c to the column it occurs in.
h. Let v be the vector: v = [1 8 8 2 1 3 9 8]. Use the unique function tocompute the total number of unique values that occur in v.

II. Programming problem: averaging images [40%]
Write a program that will average a collection of images, compute the standard deviations at each pixel, and display the results.

The images below give some examples that were generated by averaging “100 unique commemorative photographs culled from the internet” by Jason Salavon. Your program will do something similar.

There are two collects of images provided for this assignment: set1 and set2. Notice that they are all the same sizewithin a single set.
Write code to dothese things per set of images:
• Compute the average image in grayscale.
• Compute the average image in color, by averaging per RGB channel.
• Compute a matrix holding the grayscale images’ standard deviation at each pixel (i.e., X(i,j) holds the standard deviation across all the images’ gray pixel intensities at row i,column j).
• Display each of the above.

Apply these steps to the two sets, separately. In your write-up, briefly explain the results – whydo they look the way they do?

Matlab tips:
Be sure to do the necessary typecasting (uint8 and double) when working with or
displaying the images.
Some useful functions: dir, length, zeros, imshow, imread, rgb2gray, mean, std, figure, title, subplot,
The following code fragment can be used to loop through all the image files in one directory (hereassuming your images are in a sub-directory named images, with .jpg extension):
filelist = dir(‘images/*.jpg’);
for i=1:length(filelist)
imname = [‘images/’ filelist(i).name];
nextim = imread(imname);
. . .

Submission instructions: what to hand in

Prepare these files:
• Single PDF file
o Responses and code snippets for Part I questions
o Explanation and screenshot outputs for the image results in Part II
• Code files: Documented code for Part II, as .m file(s).

Submit your files through the SDSU BLACKBOARD.

No Hardcopy is required.