Saving 2D index ranges in a single variable

Let's say I have an image stored as a matrix, and want save a rectangular region of interest that I can easily isolate from the image.

One simple way to do this would be this:

% Input image
rawImage = rand(50, 100);

% Save a 6px x 3px ROI
roi_yrange = [30:32];
roi_xrange = [1:3];

% Isolate the ROI  from image and perform some kind of further
% analysis, e. g. calculate the mean value
r = rawImage(roi_yrange, roi_xrange); % isolated portion of image

avr_brightness_in_roi = mean(r, 'all')

avr_brightness_in_roi = 0.5230

That works fine, however, you always need two variables to store the ranges. Is there some way to store them in a single array, and do something like rawImage(roi_range)?

Obviously, concatenating the ranges into a matrix doesn't work if they have different lengths.

You can concatenate them into a cell array, but then indexing no longer works:

roi_range = {[30:32], [1:3]};
r = rawImage(roi_range);

Unable to use a value of type cell as an index.

Is there any way to do this, or are we stuck with having to use separate variables?

Answer 1

[Mathieu NOE]

hello

why not this :

% Input image
rawImage = rand(50, 100);
% Save a 6px x 3px ROI
% roi_xrange = [1:3];
% roi_yrange = [30:32];
xyrang = [1 3 30 32]; % first two values are x min / max, second two values are y min / max
% Isolate the ROI  from image and perform some kind of further
% analysis, e. g. calculate the mean value
% r = rawImage(roi_yrange, roi_xrange); % isolated portion of image
r = rawImage([xyrang(3):xyrang(4)], [xyrang(1):xyrang(2)]); % isolated portion of image
avr_brightness_in_roi = mean(r, 'all')

[fi]

That works, but isn't really what I was looking for – I specifically wanted to be able to do something like rawImage(roi_range) to make that indexing call short and readable.

But I guess there isn't really any way to achieve that and your answer answers my original question, so I'll mark this is solved.