I need to find the index of a 2d matrix within a 3rd array which has the lowest number of NaN values.

I have a 3D array of dimensions 942x523x365 with numerical and NaN values. I would like to find the 2D slice along the 3rd Dimension with the lowest number of NaN values.

Answer 1

[Matlab Pro]

HI @Angus

Here is a simpler example with lower dimentions..

Enjoy

M = randi([1,100],[5,10,15]);
idxNan = rand([5,10,15]) > 0.8;
M(idxNan) = nan;

% Horizontal max Nan's index
tmp  = [sum(sum(isnan(M)))];
howManyNans = squeeze(tmp(1,1,:));
maxIdx = find(howManyNans == max(howManyNans));

% Vertical max Nan's index
M1 = permute(M,[3 1 2]);
tmp  = [sum(sum(isnan(M1)))];
howManyNans = squeeze(tmp(1,1,:));
maxIdx = find(howManyNans == max(howManyNans));

[Angus]

That has worked perfectly, thank you very much!