bool computePlaneToPlaneHomography( const Eigen::MatrixXd &points1, const Eigen::MatrixXd &points2, Eigen::MatrixXd &H,
std::vector<bool> &mask, double reprojectionErrorThreshold )
{
if( points1.size() != points2.size() )
{
throw std::runtime_error( "computePlaneToPlaneHomography: Point lists must be of the same length." );
}
if( H.cols() != 3 || H.rows() != 3 )
{
H.resize( 3, 3 );
}
bool result = false;
// If we have exactly 4 points, compute the homography.
if( points1.size() == 4 )
{
result = compute4PointPlaneToPlaneHomography( points1, points2, H );
}
else
{
// Otherwise, use RANSAC to remove the outliers.
Detail::HomographyEstimator estimator(4);
estimator.setThreshold( reprojectionErrorThreshold );
result = estimator( points1, points2, H, mask );
}
return result;
}
QVariant EstimateProfileOperationCommand::makeFeeObject(const BigInt &gas, const QVariantMap &request)
{
Ethereum::Connector::GasEstimator estimator(_provider);
BigInt price = estimator.getPrice();
if(request.contains("factor"))
{
size_t factor = request["factor"].toInt();
if(factor==0)
{
price = 0;
}
else
{
price *= factor;
price /= 100;
}
}
QVariantMap result;
BigInt fee = gas * price;
result["gas"] = gas.str().c_str();
result["fee"] = fee.str().c_str();
result["price"] = price.str().c_str();
return result;
}
int main()
{
double rate = 0.0;
double vol = 0.1;
int N = 3;
int M = 1;
int b = 700;
double maturity = 1.0;
double initialSpot = 100.0;
double timeStep = (maturity / (double)N);
double m = timeStep*(rate - vol*vol / 2.0);
//std::ofstream fichier("C:\\Users\\Lucien\\Desktop\\e.txt", std::ios::out | std::ios::trunc);
for (int i = 0; i < 40; i++)
{
double initialSpots[1]{80 + (double)i};
//False ditribution in the estimator
lognormal f(m, vol*sqrt(timeStep));
lognormal g(0.0, 5.0*vol);
normalGen gGenerator(ENG(time(0)), normalDist(0.0, 5.0*vol));
RandomGenerationTool kit(f, gGenerator, g);
BlackScholesPathGenerator paths(rate, vol, N, b, M, maturity, kit);
BlackScholesRate rates(N, maturity, rate);
AmericanPutPayoff payoff(100.0, maturity, N, b, M);
Estimator estimator(paths, payoff, rates);
std::cout << estimator.computePrice(initialSpots) << std::endl;
}
//fichier.close();
system("pause");
}
int main (int argc, char *argv[]) {
ope::OPESettings settings;
settings.minTgtDepth = 0.4f;
settings.maxObjHeight = 2.5f;
ope::ObjectPoseEstimator estimator(settings);
ope::SQParameters sqParams = estimator.run();
std::cin.get();
std::cin.get();
return 0;
}
void CCtrlDialog::OnBnClickedMfcbuttonObtainucs()
{
// TODO: 在此添加控件通知处理程序代码
CDepthEstimation estimator(m_pAsmbBody);
estimator.EstimateUCS();
CMainFrame *pMain = (CMainFrame *)AfxGetMainWnd();
pMain->ShowUCSEditor();
CWnd *pWnd = GetDlgItem(IDC_MFCBUTTON_OBTAINUCS);
pWnd->EnableWindow(FALSE);
// pWnd = GetDlgItem(IDC_MFCBUTTON_DEPTHESTIM);
// pWnd->EnableWindow(TRUE);
}
TEST(ProbWelfordCovarEstimator, num_samples) {
const int n = 10;
Eigen::VectorXd q = Eigen::VectorXd::Ones(n);
const int n_learn = 10;
stan::math::welford_covar_estimator estimator(n);
for (int i = 0; i < n_learn; ++i)
estimator.add_sample(q);
EXPECT_EQ(n_learn, estimator.num_samples());
}
void CCtrlDialog::OnBnClickedMfcbuttonDepthestim()
{
// TODO: 在此添加控件通知处理程序代码
CDepthEstimation estimator(m_pAsmbBody);
estimator.Estimate();
CWnd *pWnd = GetDlgItem(IDC_MFCBUTTON_DEPTHESTIM);
pWnd->EnableWindow(FALSE);
pWnd = GetDlgItem(IDC_MFCBUTTON_RECONSTRUCTION);
pWnd->EnableWindow(TRUE);
CMainFrame *pMain = (CMainFrame *)AfxGetMainWnd();
CSLDRView * pView = (CSLDRView *)(pMain->GetActiveView());
pView->Invalidate();
}
int main(int argc, char **argv) // rosrun spheres_localization pose_estimation /home/jrhizor/Desktop/KinFuSnapshots/map.txt SIFT /camera/image_raw
{
ros::init(argc, argv, "pose_estimation");
// rosrun spheres_localization pose_estimation /home/jrhizor/Desktop/KinFuSnapshots/map.txt SIFT topicname
ROS_ASSERT(argc==4);
std::string map_file_name = argv[1];
std::string method = argv[2];
std::string camera_topic = argv[3];
// export GSCAM_CONFIG="v4l2src device=/dev/video1 ! video/x-raw-rgb,framerate=30/1 ! ffmpegcolorspace"
// rosrun gscam gscam
// load map
std::vector<InterestPoint3D> world_map = load_map(map_file_name);
ROS_INFO("Loaded map with %d points.", int(world_map.size()));
// handle map
std::vector<cv::KeyPoint> map_keypoints;
cv::Mat map_desc(world_map.size(), world_map[0].descriptor.size(), CV_32F);
ROS_INFO("Created containers for map information.");
for(unsigned int i=0; i<world_map.size(); ++i)
{
cv::KeyPoint temp_keypoint;
temp_keypoint.pt = cv::Point2f(0, 0); // value never read
for(unsigned int j=0; j<world_map[i].descriptor.size(); ++j)
{
map_desc.at<float>(i,j) = world_map[i].descriptor[j];
}
// map_position_lookup[std::make_pair(temp_keypoint.pt.x, temp_keypoint.pt.y)] =
// pcl::PointXYZ(world_map[i].x,world_map[i].y,world_map[i].z);
map_keypoints.push_back(temp_keypoint);
}
ROS_INFO("Filled containers with map information.");
PoseEstimator estimator(camera_topic);
// estimator.run(map_keypoints, map_desc, map_position_lookup, method);
estimator.run(map_keypoints, map_desc, world_map, method);
return 0;
}
Stitcher::Status Stitcher::estimateCameraParams()
{
detail::HomographyBasedEstimator estimator;
if (!estimator(features_, pairwise_matches_, cameras_))
return ERR_HOMOGRAPHY_EST_FAIL;
for (size_t i = 0; i < cameras_.size(); ++i)
{
Mat R;
cameras_[i].R.convertTo(R, CV_32F);
cameras_[i].R = R;
LOGLN("Initial intrinsic parameters #" << indices_[i] + 1 << ":\n " << cameras_[i].K());
}
bundle_adjuster_->setConfThresh(conf_thresh_);
if (!(*bundle_adjuster_)(features_, pairwise_matches_, cameras_))
return ERR_CAMERA_PARAMS_ADJUST_FAIL;
// Find median focal length and use it as final image scale
std::vector<double> focals;
for (size_t i = 0; i < cameras_.size(); ++i)
{
LOGLN("Camera #" << indices_[i] + 1 << ":\n" << cameras_[i].K());
focals.push_back(cameras_[i].focal);
}
std::sort(focals.begin(), focals.end());
if (focals.size() % 2 == 1)
warped_image_scale_ = static_cast<float>(focals[focals.size() / 2]);
else
warped_image_scale_ = static_cast<float>(focals[focals.size() / 2 - 1] + focals[focals.size() / 2]) * 0.5f;
if (do_wave_correct_)
{
std::vector<Mat> rmats;
for (size_t i = 0; i < cameras_.size(); ++i)
rmats.push_back(cameras_[i].R);
detail::waveCorrect(rmats, wave_correct_kind_);
for (size_t i = 0; i < cameras_.size(); ++i)
cameras_[i].R = rmats[i];
}
return OK;
}
TEST(ProbWelfordVarEstimator, sample_variance) {
const int n = 10;
const int n_learn = 10;
stan::math::welford_var_estimator estimator(n);
for (int i = 0; i < n_learn; ++i) {
Eigen::VectorXd q = Eigen::VectorXd::Constant(n, i);
estimator.add_sample(q);
}
Eigen::VectorXd var(n);
estimator.sample_variance(var);
for (int i = 0; i < n; ++i)
EXPECT_EQ(55.0 / 6.0, var(i));
}
void NormalEstimator::createNormalEstimator() {
TRACE(logger, "createNormalEstimator: Starting");
std::unique_ptr<pcl::NormalEstimationOMP<pcl::PointXYZRGB, pcl::Normal>> estimator(new pcl::NormalEstimationOMP<pcl::PointXYZRGB, pcl::Normal>() );
pcl::search::KdTree<pcl::PointXYZRGB>::Ptr kdTree( new pcl::search::KdTree<pcl::PointXYZRGB>() );
estimator->setSearchMethod( kdTree );
float radius = getParam<float>( normalRadius );
if(radius > 0)
estimator->setRadiusSearch( radius );
else {
int k = getParam<int>( kNN );
assert(k > 0);
estimator->setKSearch(k);
}
normalEstimator = std::move( estimator );
TRACE(logger, "createNormalEstimator: Finished");
}
int GrayWorldCommand::execute()
{
illumestimators::GrayWorldEstimator estimator(config.n, config.p, config.sigma);
SingleIlluminantPipeline::loadIlluminantEstimator(estimator, config.loadFile);
MetadataStorage storage(config.metaFiles);
Illum estimate;
double error;
if (!SingleIlluminantPipeline::runEstimator(estimate, error, estimator, config.inputFile, storage, config.verbosity)) {
return -1;
}
SingleIlluminantPipeline::saveIlluminantEstimator(estimator, config.saveFile);
return 0;
}
TEST(ProbWelfordCovarEstimator, sample_mean) {
const int n = 10;
const int n_learn = 10;
stan::math::welford_covar_estimator estimator(n);
for (int i = 0; i < n_learn; ++i) {
Eigen::VectorXd q = Eigen::VectorXd::Constant(n, i);
estimator.add_sample(q);
}
Eigen::VectorXd mean(n);
estimator.sample_mean(mean);
for (int i = 0; i < n; ++i)
EXPECT_EQ(9.0 / 2.0, mean(i));
}
TEST(ProbWelfordCovarEstimator, sample_covariance) {
const int n = 10;
const int n_learn = 10;
stan::math::welford_covar_estimator estimator(n);
for (int i = 0; i < n_learn; ++i) {
Eigen::VectorXd q = Eigen::VectorXd::Constant(n, i);
estimator.add_sample(q);
}
Eigen::MatrixXd covar(n, n);
estimator.sample_covariance(covar);
for (int i = 0; i < n; ++i)
for (int j = 0; j < n; ++j)
EXPECT_EQ(55.0 / 6.0, covar(i, j));
}
void EMPeriodicGaussian::mergeParams() {
sort(params_.begin(), params_.end(), paramComparator);
vector<bool> skip(params_.size(), 0);
// final set of parameters
vector<Param> refined;
for(int i=0; i < params_.size(); i++) {
// if this parameter has not already been merged
if(!skip[i]) {
// find longest continuous sequence of parameters
// that can be merged into a single parameter
bool hasMergedOnce = false;
vector<Param> candidates;
Param best = params_[i];
candidates.push_back(params_[i]);
for(int j=(i+1) % params_.size(); j != i; j = (j+1) % params_.size()) {
candidates.push_back(params_[j]);
Param estimate = estimator(candidates);
double squaredError = squaredIntegratedError(candidates, estimate);
if(sqrt(squaredError) < 5e-3) {
hasMergedOnce = true;
best = estimate;
// inclusive merge of all continuous indices
for(int k=i; k<=j; k++) {
skip[k] = true;
}
} else {
if(hasMergedOnce) {
break;
}
}
}
refined.push_back(best);
}
}
if(refined.size() != params_.size()) {
params_ = refined;
}
}
TEST(ProbWelfordCovarEstimator, restart) {
const int n = 10;
Eigen::VectorXd q = Eigen::VectorXd::Ones(n);
const int n_learn = 10;
stan::math::welford_covar_estimator estimator(n);
for (int i = 0; i < n_learn; ++i)
estimator.add_sample(q);
estimator.restart();
EXPECT_EQ(0, estimator.num_samples());
Eigen::VectorXd mean(n);
estimator.sample_mean(mean);
for (int i = 0; i < n; ++i)
EXPECT_EQ(0, mean(i));
}
void
RecognizerROS<PointT>::initialize (int argc, char ** argv)
{
n_.reset( new ros::NodeHandle ( "~" ) );
bool do_sift = true;
bool do_shot = false;
bool do_ourcvfh = false;
float resolution = 0.003f;
std::string models_dir, training_dir;
typename GHV<PointT, PointT>::Parameter paramGHV;
typename GraphGeometricConsistencyGrouping<PointT, PointT>::Parameter paramGgcg;
typename LocalRecognitionPipeline<flann::L1, PointT, FeatureT >::Parameter paramLocalRecSift;
typename LocalRecognitionPipeline<flann::L1, PointT, pcl::Histogram<352> >::Parameter paramLocalRecShot;
typename MultiRecognitionPipeline<PointT>::Parameter paramMultiPipeRec;
typename SHOTLocalEstimationOMP<PointT, pcl::Histogram<352> >::Parameter paramLocalEstimator;
paramLocalRecSift.use_cache_ = paramLocalRecShot.use_cache_ = true;
paramLocalRecSift.save_hypotheses_ = paramLocalRecShot.save_hypotheses_ = true;
n_->getParam ( "visualize", visualize_);
n_->getParam ( "models_dir", models_dir);
n_->getParam ( "training_dir", training_dir);
n_->getParam ( "chop_z", chop_z_ );
n_->getParam ( "do_sift", do_sift);
n_->getParam ( "do_shot", do_shot);
n_->getParam ( "do_ourcvfh", do_ourcvfh);
n_->getParam ( "knn_sift", paramLocalRecSift.knn_);
n_->getParam ( "knn_shot", paramLocalRecShot.knn_);
int normal_computation_method;
if( n_->getParam ( "normal_method", normal_computation_method) )
{
paramLocalRecSift.normal_computation_method_ =
paramLocalRecShot.normal_computation_method_ =
paramMultiPipeRec.normal_computation_method_ =
paramLocalEstimator.normal_computation_method_ =
normal_computation_method;
}
int icp_iterations;
if(n_->getParam ( "icp_iterations", icp_iterations))
paramLocalRecSift.icp_iterations_ = paramLocalRecShot.icp_iterations_ = paramMultiPipeRec.icp_iterations_ = icp_iterations;
n_->getParam ( "cg_size_thresh", paramGgcg.gc_threshold_);
n_->getParam ( "cg_size", paramGgcg.gc_size_);
n_->getParam ( "cg_ransac_threshold", paramGgcg.ransac_threshold_);
n_->getParam ( "cg_dist_for_clutter_factor", paramGgcg.dist_for_cluster_factor_);
n_->getParam ( "cg_max_taken", paramGgcg.max_taken_correspondence_);
n_->getParam ( "cg_max_time_for_cliques_computation", paramGgcg.max_time_allowed_cliques_comptutation_);
n_->getParam ( "cg_dot_distance", paramGgcg.thres_dot_distance_);
n_->getParam ( "use_cg_graph", paramGgcg.use_graph_);
n_->getParam ( "hv_clutter_regularizer", paramGHV.clutter_regularizer_);
n_->getParam ( "hv_color_sigma_ab", paramGHV.color_sigma_ab_);
n_->getParam ( "hv_color_sigma_l", paramGHV.color_sigma_l_);
n_->getParam ( "hv_detect_clutter", paramGHV.detect_clutter_);
n_->getParam ( "hv_duplicity_cm_weight", paramGHV.w_occupied_multiple_cm_);
n_->getParam ( "hv_histogram_specification", paramGHV.use_histogram_specification_);
n_->getParam ( "hv_hyp_penalty", paramGHV.active_hyp_penalty_);
n_->getParam ( "hv_ignore_color", paramGHV.ignore_color_even_if_exists_);
n_->getParam ( "hv_initial_status", paramGHV.initial_status_);
n_->getParam ( "hv_inlier_threshold", paramGHV.inliers_threshold_);
n_->getParam ( "hv_occlusion_threshold", paramGHV.occlusion_thres_);
n_->getParam ( "hv_optimizer_type", paramGHV.opt_type_);
n_->getParam ( "hv_radius_clutter", paramGHV.radius_neighborhood_clutter_);
n_->getParam ( "hv_radius_normals", paramGHV.radius_normals_);
n_->getParam ( "hv_regularizer", paramGHV.regularizer_);
n_->getParam ( "hv_plane_method", paramGHV.plane_method_);
n_->getParam ( "hv_add_planes", paramGHV.add_planes_);
int min_plane_inliers;
if(n_->getParam ( "hv_min_plane_inliers", min_plane_inliers))
paramGHV.min_plane_inliers_ = static_cast<size_t>(min_plane_inliers);
// n_->getParam ( "hv_requires_normals", r_.hv_params_.requires_normals_);
rr_.reset(new MultiRecognitionPipeline<PointT>(paramMultiPipeRec));
boost::shared_ptr < GraphGeometricConsistencyGrouping<PointT, PointT> > gcg_alg (
new GraphGeometricConsistencyGrouping<PointT, PointT> (paramGgcg));
boost::shared_ptr <Source<PointT> > cast_source;
if (do_sift || do_shot ) // for local recognizers we need this source type / training data
{
boost::shared_ptr < RegisteredViewsSource<pcl::PointXYZRGBNormal, PointT, PointT> > src
(new RegisteredViewsSource<pcl::PointXYZRGBNormal, PointT, PointT>(resolution));
src->setPath (models_dir);
src->setModelStructureDir (training_dir);
src->generate ();
// src->createVoxelGridAndDistanceTransform(resolution);
cast_source = boost::static_pointer_cast<RegisteredViewsSource<pcl::PointXYZRGBNormal, PointT, PointT> > (src);
}
if (do_sift)
{
#ifdef HAVE_SIFTGPU
boost::shared_ptr < SIFTLocalEstimation<PointT, FeatureT > > estimator (new SIFTLocalEstimation<PointT, FeatureT >());
boost::shared_ptr < LocalEstimator<PointT, FeatureT > > cast_estimator = boost::dynamic_pointer_cast<SIFTLocalEstimation<PointT, FeatureT > > (estimator);
#else
boost::shared_ptr < OpenCVSIFTLocalEstimation<PointT, FeatureT > > estimator (new OpenCVSIFTLocalEstimation<PointT, FeatureT >);
boost::shared_ptr < LocalEstimator<PointT, FeatureT > > cast_estimator = boost::dynamic_pointer_cast<OpenCVSIFTLocalEstimation<PointT, FeatureT > > (estimator);
#endif
boost::shared_ptr<LocalRecognitionPipeline<flann::L1, PointT, FeatureT > > sift_r;
sift_r.reset (new LocalRecognitionPipeline<flann::L1, PointT, FeatureT > (paramLocalRecSift));
sift_r->setDataSource (cast_source);
sift_r->setTrainingDir (training_dir);
sift_r->setFeatureEstimator (cast_estimator);
sift_r->initialize (false);
boost::shared_ptr < Recognizer<PointT> > cast_recog;
cast_recog = boost::static_pointer_cast<LocalRecognitionPipeline<flann::L1, PointT, FeatureT > > (sift_r);
std::cout << "Feature Type: " << cast_recog->getFeatureType() << std::endl;
rr_->addRecognizer (cast_recog);
}
if (do_shot)
{
boost::shared_ptr<UniformSamplingExtractor<PointT> > uniform_kp_extractor ( new UniformSamplingExtractor<PointT>);
uniform_kp_extractor->setSamplingDensity (0.01f);
uniform_kp_extractor->setFilterPlanar (true);
uniform_kp_extractor->setThresholdPlanar(0.1);
uniform_kp_extractor->setMaxDistance( 1000.0 ); // for training we want to consider all points (except nan values)
boost::shared_ptr<KeypointExtractor<PointT> > keypoint_extractor = boost::static_pointer_cast<KeypointExtractor<PointT> > (uniform_kp_extractor);
boost::shared_ptr<SHOTLocalEstimationOMP<PointT, pcl::Histogram<352> > > estimator (new SHOTLocalEstimationOMP<PointT, pcl::Histogram<352> >(paramLocalEstimator));
estimator->addKeypointExtractor (keypoint_extractor);
boost::shared_ptr<LocalEstimator<PointT, pcl::Histogram<352> > > cast_estimator;
cast_estimator = boost::dynamic_pointer_cast<LocalEstimator<PointT, pcl::Histogram<352> > > (estimator);
boost::shared_ptr<LocalRecognitionPipeline<flann::L1, PointT, pcl::Histogram<352> > > local;
local.reset(new LocalRecognitionPipeline<flann::L1, PointT, pcl::Histogram<352> > (paramLocalRecShot));
local->setDataSource (cast_source);
local->setTrainingDir(training_dir);
local->setFeatureEstimator (cast_estimator);
local->initialize (false);
uniform_kp_extractor->setMaxDistance( chop_z_ ); // for training we do not want this restriction
boost::shared_ptr<Recognizer<PointT> > cast_recog;
cast_recog = boost::static_pointer_cast<LocalRecognitionPipeline<flann::L1, PointT, pcl::Histogram<352> > > (local);
std::cout << "Feature Type: " << cast_recog->getFeatureType() << std::endl;
rr_->addRecognizer(cast_recog);
}
boost::shared_ptr<GHV<PointT, PointT> > hyp_verification_method (new GHV<PointT, PointT>(paramGHV));
boost::shared_ptr<HypothesisVerification<PointT,PointT> > cast_hyp_pointer = boost::static_pointer_cast<GHV<PointT, PointT> > (hyp_verification_method);
rr_->setHVAlgorithm( cast_hyp_pointer );
rr_->setCGAlgorithm( gcg_alg );
vis_pc_pub_ = n_->advertise<sensor_msgs::PointCloud2>( "sv_recogniced_object_instances", 1 );
recognize_ = n_->advertiseService ("sv_recognition", &RecognizerROS::recognizeROS, this);
it_.reset(new image_transport::ImageTransport(*n_));
image_pub_ = it_->advertise("sv_recogniced_object_instances_img", 1, true);
}
void ModuleServices::estimateSongLength()
{
SongLengthEstimator estimator(&module);
estimatedSongLength = estimator.estimateSongLengthInSeconds();
}
int ImageReranker::cvFindHomography( const CvMat* objectPoints, const CvMat* imagePoints,
CvMat* __H, int method, double ransacReprojThreshold,
CvMat* mask )
{
const double confidence = 0.995;
const int maxIters = 400;
const double defaultRANSACReprojThreshold = 3;
bool result = false;
Ptr<CvMat> m, M, tempMask;
double H[9];
CvMat matH = cvMat( 3, 3, CV_64FC1, H );
int count;
CV_Assert( CV_IS_MAT(imagePoints) && CV_IS_MAT(objectPoints) );
count = MAX(imagePoints->cols, imagePoints->rows);
CV_Assert( count >= 4 );
if( ransacReprojThreshold <= 0 )
ransacReprojThreshold = defaultRANSACReprojThreshold;
m = cvCreateMat( 1, count, CV_64FC2 );
cvConvertPointsHomogeneous( imagePoints, m );
M = cvCreateMat( 1, count, CV_64FC2 );
cvConvertPointsHomogeneous( objectPoints, M );
if( mask )
{
CV_Assert( CV_IS_MASK_ARR(mask) && CV_IS_MAT_CONT(mask->type) &&
(mask->rows == 1 || mask->cols == 1) &&
mask->rows*mask->cols == count );
}
if( mask || count > 4 )
tempMask = cvCreateMat( 1, count, CV_8U );
if( !tempMask.empty() )
cvSet( tempMask, cvScalarAll(1.) );
CvHomographyEstimator estimator(4);
if( count == 4 )
method = 0;
assert(method == CV_RANSAC);
result = estimator.runRANSAC( M, m, &matH, tempMask, ransacReprojThreshold, confidence, maxIters);
if( result && count > 4 )
{
icvCompressPoints( (CvPoint2D64f*)M->data.ptr, tempMask->data.ptr, 1, count );
count = icvCompressPoints( (CvPoint2D64f*)m->data.ptr, tempMask->data.ptr, 1, count );
M->cols = m->cols = count;
if( method == CV_RANSAC )
estimator.runKernel( M, m, &matH );
estimator.refine( M, m, &matH, 10 );
}
if( result )
cvConvert( &matH, __H );
if( mask && tempMask )
{
if( CV_ARE_SIZES_EQ(mask, tempMask) )
cvCopy( tempMask, mask );
else
cvTranspose( tempMask, mask );
}
return (int)result;
}