Do I have to loop through every pixel when I want to mask one 2D Dataset with another?
float[] dataRead = (float[]) data.read();
float[] maskRead = (float[]) mask.read();
for (int i = 0; i < dims[0] * dims[1] ; ++i) {
if (maskRead[i] == 1) {
/* 1 * dataRead[i] */
}
else {
dataRead[i] = fillvalue;
/* ...or 0 * dataRead[i] */
}
}
data.write(dataRead);
Or is there some sort of function/technique which will multiply all of the corresponding pixels together?
Richard,
There is no such functionality in HDF5.
Said this, you may wish to look at point selection in HDF5 http://www.hdfgroup.org/HDF5/Tutor/selectc.html and HDF5 region references http://www.hdfgroup.org/HDF5/Tutor/reftoreg.html.
Some examples are also available from http://www.hdfgroup.org/ftp/HDF5/examples/examples-by-api/api18-c.html.
What you can do is to define a point selection that corresponds to all "1" in a dataset with the mask values. Then use this selection to read data from a dataset assuming that a dataset with mask values and the dataset to be read have the same dimensionality (rank and dimension sizes) .
Elena
···
On Aug 12, 2010, at 12:53 PM, Richard Strub wrote:
Do I have to loop through every pixel when I want to mask one 2D Dataset with another?
float[] dataRead = (float[]) data.read();
float[] maskRead = (float[]) mask.read();
for (int i = 0; i < dims[0] * dims[1] ; ++i) {
if (maskRead[i] == 1) {
/* 1 * dataRead[i] */
}
else {
dataRead[i] = fillvalue;
/* ...or 0 * dataRead[i] */
}
}
data.write(dataRead);
Or is there some sort of function/technique which will multiply all of the corresponding pixels together?
_______________________________________________
Hdf-forum is for HDF software users discussion.
Hdf-forum@hdfgroup.org
http://mail.hdfgroup.org/mailman/listinfo/hdf-forum_hdfgroup.org
