# Selection of priority data

2 views (last 30 days)
Giuseppe on 26 Dec 2014
Answered: Giuseppe on 27 Dec 2014
Hello,
I'm trying to solve a problem. I have a vector of data such as:
data=[3;4;8;NaN;NaN;NaN;7;4;3;NaN;NaN;2;3;NaN;9]
length of data can be variable and NaN value phases can happen everywhere within a vector.
I need to find the first NaN value and count the number of next NaN. However, I do not want to take into account any other NaN value after the first portion of NaN.
In this case my result would be: First NaN at raw 4. Three NaN values. I know how to use the function find, indeed this is the code I used as far:
findNaN = find(isnan(data));
The problem is that in that way I will get the others NaN too that I do not want to locate.
Any idea?
Thanks, Giuseppe

Shoaibur Rahman on 27 Dec 2014
data=[3;4;8;NaN;NaN;NaN;7;4;3;NaN;NaN;2;3;NaN;9];
findNaN = find(isnan(data));
first = findNaN(1);
numberNaN = 1;
for k = first+1:length(data)
if isnan(data(k))
numberNaN = numberNaN+1;
else break
end
end
first, numberNaN

Image Analyst on 27 Dec 2014
If you're lucky enough to own the Image Processing Toolbox, it's easy. Just find the nan's, label then, and count the number in the first group. Here is the code:
labeledVector = bwlabel(isnan(data)) % Find nan's and label.
numberOfNansInFirstGroup = sum(labeledVector == 1) % Count the # of elements in group #1

Giuseppe on 27 Dec 2014
thanks for helping to improve my code