I need to do Quantization to an RGB image in matlab based on 3 slider values as shown below. The values cant be decimals, just integers 0 - 8

rgbBitDepath: A 1x3 row vector; each of the three vector elements represents bit depth of the RGB channel. For example, if the user wants to reduce a 24-bit image (i.e., (8-bit, 8-bit, 8-bit) to (3-bit, 3-bit, 2-bit)), the argument should be [3,3,2], whichrepresents R:3bits, G:3bits, B:2bits. Each bit depth value for each RGB channel should be less than 8-bit.

I got it all working except the quantization. How can I impliment this?