How to get the position of minimum intensity in an image using MATLAB programming

Question

I have a image with me. I would like to obtain and mark the position having minimum intensity in that image using MATLAB programming. I would also like to get the value of minimum intensity

Answer 1

To correctly identify where in the image the minimum intensity is, combine a call with min and find. As such, if A is your image, do the following:

min_val = min(A(:));
[row,col] = find(A == min_val), 1);

min_val will contain the minimum intensity in your image. What is important is unrolling the matrix A to a single column vector by A(:). This way, you are ensuring that min operates along all of the pixel values at the same time. In the above code, row and col will contain the row and column locations of where minimum intensity was found. The intricacy with this is that if you have more than one pixel that shares the same minimum intensity, find will return all locations that share this same intensity. If you just want one of them, append 1 as the second parameter to find.

If you want to mark where in the image the minimum intensity is, that's simply done by indexing:

A(row,col) = 255;

This will set the location of where the minimum intensity is located to be white or 255.

C.Colden's code is also a good attempt, but it will not work if the input is not a vector (i.e. a matrix or array that has more than two dimensions) as min will operate along the first non-singleton dimension. This is why you need to unroll the image into a single vector as in my code above.

If we input a matrix into min, in your case, min will find the minimum along each column. If you want, you would have to call min twice to ascertain the row location by applying min on the column found by the first min call but I find the above code more readable.

However, for academic purposes, this is what you would do if you wanted to only use min:

[min_col, rows] = min(A); %// Find the minimum among all columns as well as which row for each column gave the minimum
[min_val, col] = min(min_col); %// Now, for each of the minimum values, figure out which column gave us the global minimum as well as the minimum itself
row = rows(col); %// Now find the row location of this global minimum

As you can see, I would rather stick with the first attempt with find and min.... it's just more readable.

Answer 2

You can load your image using imread(http://ch.mathworks.com/help/matlab/ref/imread.html). Afterwards you can treat the image like a matrix. So if "A" is your image, you can use max (http://ch.mathworks.com/help/matlab/ref/max.html) or min to find the values using Min_Value = min(A(:)). Afterwards you can search for the location of the Min_Value in your matrix using [row,column] = find(A == Min_Value)

Example:

A = [3,2,1;4,5,20;7,8,9];
Min_Value = min(A(:));
[row,column]=find(A == Min_Value);