Hi everybody,
until now I resampled an Irregular2DSet resembling a terrain surface, for
achieving a
profile line the following way, which worked very well:
RealType x2 = RealType.getRealType("x");
RealType y2 = RealType.getRealType("y");
RealType height2 = RealType.getRealType("height");
RealTupleType xy2 = new RealTupleType(x2, y2);
FunctionType terrain_type2 = new FunctionType(xy2,height2);
Irregular2DSet set2 = new Irregular2DSet(
xy2,
new float[][] { eastValues,northValues });
FlatField terrain2 = new FlatField(terrain_type2,set2);
terrain2.setSamples(new float[][] { heightValues});
// Why 2000 points? Try and error, it was okay for
// dataset with max. 50 points.
// The more points, the better fit of the profile to the terrain.
double[][] lineCoordinates = new double[2][2000];
double j = -1;
for (int i = 0; i < 2000; i++) {
lineCoordinates[0][i] = linePointCalc.x +j * lineVector.x;
lineCoordinates[1][i] = linePointCalc.y +j * lineVector.y;
j = j + 0.001;
}
Gridded2DDoubleSet gridded2DSet
new Gridded2DDoubleSet(xy2,
lineCoordinates, 2000);
FlatField line =(FlatField) terrain2.resample(
gridded2DSet,
Data.WEIGHTED_AVERAGE,
Data.NO_ERRORS);
System.out.println(line.toString());
Now I have a grid data set with a lot more points so I want to use a
Linear2DSet as my original data set instead of the Irregular2DSet
(which is very slow now, because I have many more terrain points):
RealType eastV =RealType.getRealType("eastValues");
RealType northV =RealType.getRealType("northValues");
RealType heightV =RealType.getRealType("heightValues");
domain_tuple = new RealTupleType(eastV, northV);
func_en_h = new FunctionType(domain_tuple,heightV);
domain_set = new Linear2DSet(
domain_tuple,
northMin,
northMax,
nRows,
eastMin,
eastMax,
nCols);
double[][] flat_samples = new double[1][nCols *nRows];
for (int c = 0; c < (nRows); c++) {
for (int r = 0; r < (nCols); r++) {
// dgm 1 x y z
flat_samples[0][c * nCols + r] =height[c * nCols + r];
}
}
vals_ff = new FlatField(func_en_h, domain_set);
vals_ff.setSamples(flat_samples, false);
// Why 2000 points? Try and error, it was okay for
// dataset with max. 50 points.
// The more points, the better fit of the profile to the terrain.
double[][] lineCoordinates = new double[2][2000];
double j = -1;
for (int i = 0; i < 2000; i++) {
lineCoordinates[0][i] = linePointCalc.x + j *lineVector.x;
lineCoordinates[1][i] = linePointCalc.y + j *lineVector.y;
j = j + 0.001;
}
Gridded2DDoubleSet gridded2DSet
new Gridded2DDoubleSet(domain_tuple,
lineCoordinates, 2000);
FlatField line =(FlatField) vals_ff.resample(
gridded2DSet,
Data.WEIGHTED_AVERAGE,
Data.NO_ERRORS);
System.out.println(line.toString());
The problem is, that I only get missing values in the FlatField line. What
am I doing wrong?
Thanks for your help.
Greetings Desiree
oooooooooooooooooooooooooooooooooooooooooooooooo
Desiree Hilbring
Institut fuer Photogrammetrie und Fernerkundung
Universitaet Karlsruhe, Germany
email: hilbring@xxxxxxxxxxxxxxxxxxxx
# 0721 6083676
oooooooooooooooooooooooooooooooooooooooooooooooo
