private void findNeighbors() {
// find the nearest-neighbor for each point in the cloud
nn.setPoints(usedNnData, listPointVector.toList());
for (int i = 0; i < listPointVector.size; i++) {
// find the nearest-neighbors
resultsNN.reset();
double[] targetPt = usedNnData.get(i);
// numNeighbors+1 since the target node will also be returned and is removed
nn.findNearest(targetPt, maxDistanceNeighbor, numNeighbors + 1, resultsNN);
PointVectorNN p = listPointVector.get(i);
// save the results
p.neighbors.reset();
for (int j = 0; j < resultsNN.size; j++) {
NnData<PointVectorNN> n = resultsNN.get(j);
// don't add the point to its own list of neighbors list
if (n.point != targetPt) {
p.neighbors.add(n.data);
}
}
}
}
/** Converts points into a format understood by the NN algorithm and initializes it */
public void process(List<Point3D_F64> cloud) {
nn.init(3);
// swap the two lists to recycle old data and avoid creating new memory
Stack<double[]> tmp = unusedNnData;
unusedNnData = usedNnData;
usedNnData = tmp;
// add the smaller list to the larger one
unusedNnData.addAll(usedNnData);
usedNnData.clear();
// convert the point cloud into the NN format
for (int i = 0; i < cloud.size(); i++) {
Point3D_F64 p = cloud.get(i);
double[] d;
if (unusedNnData.isEmpty()) {
d = new double[3];
} else {
d = unusedNnData.pop();
}
d[0] = p.x;
d[1] = p.y;
d[2] = p.z;
usedNnData.add(d);
}
// declare the output data for creating the NN graph
listPointVector.reset();
for (int i = 0; i < cloud.size(); i++) {
PointVectorNN p = listPointVector.grow();
p.reset();
p.p = cloud.get(i);
p.index = i;
}
findNeighbors();
}