Find index of element closest to other index

I have two binary vectors, b1 and b2, with values of either 0 or 1. b1 has always fewer true values or ones than b2. The true values rarely overlap, i.e. have the same index.

What I want to do is to use the index of when b1 is 1 and set the closest true value or 1 in b2 to false or 0.

Basically I want to remove duplicates from b2, but if they don't overlap I want to remove closest true value.

Answer 1

[Stephen23]

>> B1 = [0,0,1,0,0,0,1,0,0,0,0,1,0,0,0,1]
B1 =
   0   0   1   0   0   0   1   0   0   0   0   1   0   0   0   1
>> B2 = [1,1,0,1,1,0,0,1,1,1,0,0,0,0,1,1]
B2 =
   1   1   0   1   1   0   0   1   1   1   0   0   0   0   1   1
>> X1 = find(B1);
>> X2 = find(B2);
>> X3 = interp1(X2,X2,X1,'nearest');
>> B2(X3) = 0
B2 =
   1   1   0   0   1   0   0   0   1   0   0   0   0   0   1   0

Note that the definition of "nearest" is open to interpretation: what happens if there are two 1's in B2 exactly the same distance from a 1 in B1? (the first 1 in B1 is an example of this)

You might need to check the edge-cases as well... probably some fine-tuning is required.