Using 2 nested for loops: Evaluate each pixel's R,G,B value : If the pixel's R value is greater than the user's threshold value for Red, set that value = to threshold instead

The underlined part is what I need help with. I have the threshold values and the image part down, I am just not sure how to do this for loop.

Using 2 nested for loops:

a. Evaluate each pixel's R,G,B value

b. If the pixel's R value is greater than the user's threshold value for Red, set that value equal to the threshold instead

c. If the pixel's R value is less than or equal to the threshold value for Red, leave it alone (nothing changes)

d. Do this for each pixel's Green and Blue intensity values

For reference, this is the first part of the question that I already have done. I just need the loop part.

a) Bring in the data associated with the image "zippy.jpg" available on Brightspace.

b) Prompt the user to enter 3 values:

a. A threshold value for the red channel (must be an integer value between 0-255)

b. A threshold value for the green channel (must be an integer value between 0-255)

c. A threshold value for the blue channel (must be an integer value between 0-255)

Any help is appreciated, I still struggle to understand loops.

Thank you very much!

J

Answer 1

[DGM]

In practice you wouldn't use a loop at all.

A = imread('peppers.png');

maxthresh = uint8([100 150 200]); % [R G B]

B = min(A,permute(maxthresh,[1 3 2]));

imshow(B)

Though if you want to do it in a loop, you can. It'll just be unnecessarily slow.

A = imread('peppers.png');

maxthresh = [100 150 200]; % [R G B]
maxthresh = uint8(permute(maxthresh,[1 3 2])); 

[h,w,~] = size(A);
for r = 1:h
    for c = 1:w
        A(r,c,:) = min(A(r,c,:),maxthresh);
    end
end

imshow(A)

[JA]

That is exactly what I needed, Thank you very much! That is good information to know, using the loop is just what my professor is requiring for this assignment.

Thank you again!

[Image Analyst]

You should have said it was homework before. Then he could have given you hints instead of a full solution. I'm sure your professor won't allow you to turn in @DGM's solution as your own.

[DGM]

Honestly, if it were me, I'd write the operation in the vectorized form and then include a token set of nested loops that contains nothing but a comment to the TA that using loops for such a thing is unnecessary and shouldn't be the focus of the assignment if the goal is to teach students how to use MATLAB. I doubt the TA would disagree (if they actually looked at the code at all).

Disclaimer: Just because I like being a smartass doesn't mean it's universally recommendable. Understand the people you work with and use your discretion.

[JA]

@Image Analyst I appologize for not saying it was homework. This was a small part of a much longer problem that I was stuck on. I just did not know how to order things in the loop. All the other code for this problem was already written. @DGM Thank you for your help, I feel like I understand the process more.

also, would either of you be able to explain symbolic expressions and function handels? I don't understand what function handels mean or how they work/what they do. In my notes we have an example

syms x

exp1 = x^3 + 3*x^2 -4*x +3;

exp2 = 1/x

%find derivative of exp1

mydiff = diff(exp1)

%find the integral of exp2

myint = int(exp2)

%evaluate the integral of exp1 from 1 to 3 with quad

f = @(x) log(x)

answer = quad(f, 1, 3)

I just don't understand what the @(x) does or how it is different from f(x).

I was out sick during that lecture, so I am a little lost on the notes. Any explanation is appreciated.

[DGM]

Think of this line

% function f() is a function of the input argument x
% and its output is log(x)
f = @(x) log(x);

as basically any regular old MATLAB function

function fx = myfunction(x) % define output and input arguments and function name
    fx = log(x); % assign a value to the output arg based on the input arg
end

The x that's used in either case is local to the function itself. You could write

f = @(a) log(a);

and it would mean the same thing, so don't confuse the x on that line with the symbolic variable x that's used on prior lines.

Either way, you're just creating a function that accepts numeric inputs and produces numeric outputs. When you call quad(), it will internally call f() in order to evaluate the function and approximate its integral.

[JA]

Thank you for that explanation. Seeing it that way makes much more sense to me. Thank you again for all of your help!