Given a center, sum adjacent pixels

Hi, suppose I have a matrix A and I have a center (for example (10.3,14.6)). Now I want to calculate the value of the center in this way

value =  w1*A(10,14) +  w2*A(10,15) +  w3*A(11,14) +  w4*A(11,15)

w1, w2, w3, w4 are the values of the weigths that are proportional to the distance of the pixel from the center.

What is the most correct way to evaluate w1, w2, w3, w4? I want that the weigth of the pixel that is closer to the center higher and the weigth of the pixel that is farther to the center lower and so on.

My hyphotesis is something like this:

w1=0.7 * 0.4 ; w2=0.7*0.6; w3=0.3*0.4; w4= 0.3*0.6

Thanks

[Benjamin Thompson]

Can you add more to your description with sample values you would like to see for w1, w2, w3, w4 as part of your example?

[MementoMori]

@Benjamin Thompson

Hi, thanks for the answer I have added my hyphotesis and a description

[dpb]

Are you saying you want to solve for a set of weights that will produce the given value from the four or recompute a revised "center" by weight the four values?

[MementoMori]

@dpb

Hi, thanks for the answer.

No, I want to calculate a sort of value of the center, making a weighted sum of the pixels around the center. The weights have to be proportional to the distance of the pixel from the center.

[dpb]

NB the weights above sum to 2 which will double the value...and "proportional" how (linear, quadratic, exponential, ...) and by what measure of "distance"?

[MementoMori]

@dpb

the sum of the weights is 1. 0.7*0.4 + 0.7*0.6 + 0.3*0.4 + 0.3*0.6 =1

I think linear proportional and I don't know for distance, what is the most suitable?

Answer 1

[Matt J]

Use interp2:

value=interp2(A,14.3,10.6)