I'm struggling to understand the use of a 1-pixel box-blur. To blur the image, why wouldn't we have to use a bigger kernel? Isn't a pixel already the most granular image unit? In my current understanding, a 1 x 1 kernel would simply scale the image but I know that I must be wrong.
I'm struggling to understand the use of a 1-pixel box-blur. To blur the image, why wouldn't we have to use a bigger kernel? Isn't a pixel already the most granular image unit? In my current understanding, a 1 x 1 kernel would simply scale the image but I know that I must be wrong.