generic-library/opencv
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Pick centers in KMeans++ with a probability proportional to their distance^2, instead of simple distance, to previous centers

Browse Source
This commit is contained in:
Pierre-Emmanuel Viel 2013-12-17 12:51:58 +01:00
parent 459e7d4a80
commit 45e0e5f8e9
1 changed files with 11 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

13
modules/flann/include/opencv2/flann/hierarchical_clustering_index.h
View File

@ -210,8 +210,11 @@ private:
assert(index >=0 && index < n); assert(index >=0 && index < n);
centers[0] = dsindices[index]; centers[0] = dsindices[index];
// Computing distance^2 will have the advantage of even higher probability further to pick new centers
// far from previous centers (and this complies to "k-means++: the advantages of careful seeding" article)
for (int i = 0; i < n; i++) { for (int i = 0; i < n; i++) {
closestDistSq[i] = distance(dataset[dsindices[i]], dataset[dsindices[index]], dataset.cols); closestDistSq[i] = distance(dataset[dsindices[i]], dataset[dsindices[index]], dataset.cols);
closestDistSq[i] *= closestDistSq[i];
currentPot += closestDistSq[i]; currentPot += closestDistSq[i];
} }
@ -237,7 +240,10 @@ private:
// Compute the new potential // Compute the new potential
double newPot = 0; double newPot = 0;
for (int i = 0; i < n; i++) newPot += std::min( distance(dataset[dsindices[i]], dataset[dsindices[index]], dataset.cols), closestDistSq[i] ); for (int i = 0; i < n; i++) {
DistanceType dist = distance(dataset[dsindices[i]], dataset[dsindices[index]], dataset.cols);
newPot += std::min( dist*dist, closestDistSq[i] );
}
// Store the best result // Store the best result
if ((bestNewPot < 0)||(newPot < bestNewPot)) { if ((bestNewPot < 0)||(newPot < bestNewPot)) {
@ -249,7 +255,10 @@ private:
// Add the appropriate center // Add the appropriate center
centers[centerCount] = dsindices[bestNewIndex]; centers[centerCount] = dsindices[bestNewIndex];
currentPot = bestNewPot; currentPot = bestNewPot;
for (int i = 0; i < n; i++) closestDistSq[i] = std::min( distance(dataset[dsindices[i]], dataset[dsindices[bestNewIndex]], dataset.cols), closestDistSq[i] ); for (int i = 0; i < n; i++) {
DistanceType dist = distance(dataset[dsindices[i]], dataset[dsindices[bestNewIndex]], dataset.cols);
closestDistSq[i] = std::min( dist*dist, closestDistSq[i] );
}
} }
centers_length = centerCount; centers_length = centerCount;