Help using the bitand function

Hi, I'm trying to use the bitand function to drop the two least significant bits off of an image (basically forcing an 8-bit image to be displayed as 6 bit) using the bitand function. the code i'm using goes as

bubbles6 = bitand(bubbles , 252) ;

with 252 being the new assumed bit total (11111100 in binary since the two lowest are dropped), but the function keeps returning the erros message

'Error using bitand

Double inputs must have integer values in the range of ASSUMEDTYPE.'

Could someone explain what this means and why?

[Walter Roberson]

what is class(bubbles)? And does it contain any negative values?

[Anthony Koning]

Bubbles is the name of the image that I am trying to reduce the bit value of. The actual image (using the whos command) is a unit8 class grayscale image, and as such, should contain no negative values.

[Walter Roberson]

Works when I test it in r2022b.

try

bubbles6 = bitand(bubbles , 252, 'uint8') ;

and please confirm

min(bubbles, [], 'all')
max(bubbles, [], 'all')

[Anthony Koning]

Unfortunaetly, it does not work on my current build of Matlab (2021a). Inputting the code you sent just leads to an error message of 'Operands to BIT Ops must be numeric.'

[Walter Roberson]

I just tried in R2021a pre-release that I happen to have installed on a drive.

bubbles = imread('flamingos.jpg');
bubbles6 = bitand(bubbles , 252, 'uint8') ;

min(diff(unique(ubbles6)))

worked fine.

In your code, put in a breakpoint at the bitand line. Then what shows up for

which bitand(bubbles)
min(bubbles(:))
max(bubbles(:))