Subscript indices must either be real positive integers or logicals.

I get returned from a function I call the "pixel location".

At that location, what I try to do is:

A(pixel_location(1),pixel_location(2))=1;

But, I get the following error:

Subscript indices must either be real positive integers or logicals.

Why is that? How can I solve this issue?

Answer 1

[Star Strider]

Your ‘pixel_location’ vector likely contains values that are not integers greater than 0. Negative values, non-integer values, and zero are not valid subscripts in MATLAB array indexing.

[med-sweng]

Thanks for your reply. Would adding a conditional statement to check for such values solve the issue?

[Star Strider]

A conditional statement might be able to check to be certain that particular pairs from ‘pixel_location’ are positive integers, but what does that do to your code?

I would see what ‘pixel_location’ contains, and correct it if possible. For instance if it contains integers with negative values or zero, adding a positive offset to be the minimum of ‘pixel_location’+1 would correct the problem. (You may have to make appropriate changes elsewhere in your code to account for that.) If it contains non-integer values, you would have to make other changes (perhaps scaling and rounding) to make them work as indices.

Since I don’t know what ‘pixel_location’ contains, I cannot be more specific than that.

Answer 2

[Erik S.]

Check the values of pixel_location(1) and pixel_location(2). I guess you find either the value 0 or a decimal value. Indexes have to be integers.

[med-sweng]

Thanks for your reply. Would adding a conditional statement to check for such values solve the issue?

[Erik S.]

Do you know the values of pixel_location? Will it change or is it constant?

[med-sweng]

I have an iteration, for each iteration they will have different values.

[Erik S.]

If you can formulate a conditional statement which is updated each iteration as well it could work. But if you don't know which values are correct for pixel_location in each iteration it seems difficult.

Answer 3

[Image Analyst]

Try to round the values (if they might be fractional values) and clip the values to between your image limits:

[rows, columns, numberOfColorChannels] = size(yourImage);
pixel_location = round(pixel_location) % Round.
pixel_location(1) = min(pixel_location(1), rows);     % Don't allow more than rows.
pixel_location(1) = max(pixel_location(1), 1);        % Don't allow less than 1.
pixel_location(2) = min(pixel_location(2), columns);  % Don't allow more than columns.
pixel_location(2) = max(pixel_location(2), 1);        % Don't allow less than 1.