Identifying blocks of data from a cell array having the same header

I usually use xlsread to load large excel datasheets for further manipulation. The excel worksheet looks somewhat like this:

Sometimes I need to isolate all the columns that have the same header, i.e., in the above case I need to separate all columns under the Orange, Red and Green headers, and create new arrays from those. Is it possible to have an array just composed of the headers, and maybe another array giving the starting and the ending number of the columns? The Output should be something like:

HeaderArray= {'Day 12' 'Night 24' 'Morning 3'}
HeaderStart=[1 5 11]
HeaderEnd=[4 10 13]

[goerk]

Can you provide an example input?

Answer 1

[Guillaume]

Assuming that the runs of identical headers are always continuous

header = repelem({'Day 12', 'Night 24', 'Morning 3'}, [4, 6, 3]);  %construct demo data

[HeaderArray, ~, id] = unique(header, 'stable');
locs = find(diff([0; id; 0]) ~= 0);
HeaderStart = locs(1:end-1);
HeaderEnd = locs(2:end) - 1;

[Saeid]

Thanks Guillaume!