Mask unwanted data with zeros
10 Ansichten (letzte 30 Tage)
Ältere Kommentare anzeigen
bugguts99
am 3 Jun. 2011
Beantwortet: Madhava Teja Munagala
am 25 Feb. 2018
I have a matrix of dimensions 9634 x 80. I have a number of quality flags where I would like to mask the 'bad' data with zeros. I have created a mask where:
mask = zeros(size(data));
I have identified parts of this matrix where this occurs (for example, row and columns using:
[r, c] = find(idx > 0);
How then do I clear the contents of entire row and not just those cells where the quality flag is located?
0 Kommentare
Akzeptierte Antwort
Walter Roberson
am 3 Jun. 2011
After you have done the above,
data(r,:) = 0;
if you want to clear it in the data. If you want to instead build a mask in which 0 indicates data to be ignored and 1 indicates data to keep, then start with
mask = ones(size(data));
and then to mask off the rows listed in r,
mask(r,:) = 0;
Weitere Antworten (2)
Madhava Teja Munagala
am 24 Feb. 2018
Bearbeitet: Walter Roberson
am 24 Feb. 2018
mask = zeros(size(i));
mask(25:end-25;25:end-25)=1;
imshow(mask);
May i know wat exactly 25:end-25 specifies
1 Kommentar
Walter Roberson
am 24 Feb. 2018
end designates the size of the array in that dimension. So M(25:end-25;25:end-25) is the same as M(25:size(M,1)-25,25:size(M,2)-25)
What the code is doing is creating a matrix of 0 and then painting most of the inside with 1. The part that is left as 0 is a border that is 24 pixels wide from the left, 25 pixels wide from the right, 24 pixels high from the top, and 25 pixels wide from the bottom.
It is fairly likely that the person who coded it thought that they were leaving a border of 25 at the top and left, but did not use the correct index for that, and further it is likely that they only accidentally got the 25 on the right and bottom, that if they had corrected the border that they would have made a mistake on the bottom and right.
If you want a border N wide then the index should be N+1:end-N
Siehe auch
Kategorien
Mehr zu Author Block Masks finden Sie in Help Center und File Exchange
Produkte
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!