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

Prefixed constants in flann with FLANN_ to prevent clashes with constants from other includes, closes bug #890

Browse Source
This commit is contained in:
Marius Muja
2011-02-16 08:42:52 +00:00
parent 0725a31e5a
commit 53e6bab678
11 changed files with 82 additions and 92 deletions
Download Patch File Download Diff File Expand all files Collapse all files

16
modules/flann/include/opencv2/flann/autotuned_index.h
View File

@@ -44,7 +44,7 @@ namespace cvflann
struct AutotunedIndexParams : public IndexParams {
AutotunedIndexParams( float target_precision_ = 0.8, float build_weight_ = 0.01,
float memory_weight_ = 0, float sample_fraction_ = 0.1) :
IndexParams(AUTOTUNED),
IndexParams(FLANN_INDEX_AUTOTUNED),
target_precision(target_precision_),
build_weight(build_weight_),
memory_weight(memory_weight_),
@@ -55,8 +55,6 @@ struct AutotunedIndexParams : public IndexParams {
float memory_weight; // index memory weighting factor
float sample_fraction; // what fraction of the dataset to use for autotuning
flann_algorithm_t getIndexType() const { return algorithm; }
void print() const
{
logger().info("Index type: %d\n",(int)algorithm);
@@ -123,13 +121,13 @@ public:
logger().info("----------------------------------------------------\n");
flann_algorithm_t index_type = bestParams->getIndexType();
switch (index_type) {
case LINEAR:
case FLANN_INDEX_LINEAR:
bestIndex = new LinearIndex<ELEM_TYPE>(dataset, (const LinearIndexParams&)*bestParams);
break;
case KDTREE:
case FLANN_INDEX_KDTREE:
bestIndex = new KDTreeIndex<ELEM_TYPE>(dataset, (const KDTreeIndexParams&)*bestParams);
break;
case KMEANS:
case FLANN_INDEX_KMEANS:
bestIndex = new KMeansIndex<ELEM_TYPE>(dataset, (const KMeansIndexParams&)*bestParams);
break;
default:
@@ -211,7 +209,7 @@ public:
*/
virtual flann_algorithm_t getType() const
{
return AUTOTUNED;
return FLANN_INDEX_AUTOTUNED;
}
private:
@@ -347,7 +345,7 @@ private:
for (size_t i=0; i<ARRAY_LEN(maxIterations); ++i) {
for (size_t j=0; j<ARRAY_LEN(branchingFactors); ++j) {
kmeansCosts[cnt].second.centers_init = CENTERS_RANDOM;
kmeansCosts[cnt].second.centers_init = FLANN_CENTERS_RANDOM;
kmeansCosts[cnt].second.iterations = maxIterations[i];
kmeansCosts[cnt].second.branching = branchingFactors[j];
@@ -569,7 +567,7 @@ private:
float searchTime;
float cb_index;
if (bestIndex->getType() == KMEANS) {
if (bestIndex->getType() == FLANN_INDEX_KMEANS) {
logger().info("KMeans algorithm, estimating cluster border factor\n");
KMeansIndex<ELEM_TYPE>* kmeans = (KMeansIndex<ELEM_TYPE>*)bestIndex;
float bestSearchTime = -1;