pcl :: StatisticalOutlierRemoval无法在sor.filter（）运行时运行

Question

。该程序看起来像被绞死。 并且没有错误。

我认为这些pcd文件有问题，因为它可以很好地与PCL提供的pcd数据一起使用。我的pcd文件是从Kinect2保存的。但我找不到它们之间的区别。

boost::shared_ptr<pcl::PointCloud<pcl::PointXYZ> > cloud(new pcl::PointCloud<pcl::PointXYZ>); 
boost::shared_ptr<pcl::visualization::PCLVisualizer> viewer(
new pcl::visualization::PCLVisualizer("Point Cloud Viewer")); 
viewer->setCameraPosition(0.0, 0.0, -2.5, 0.0, 0.0, 0.0); 
pcl::io::loadPCDFile("table_scene_lms400.pcd",*cloud); 
//pcl::io::loadPCDFile("test.pcd",*cloud); 
pcl::StatisticalOutlierRemoval<pcl::PointXYZ> sor; 
sor.setMeanK(50); 
sor.setStddevMulThresh(1); 
pcl::PointCloud <pcl::PointXYZ>::Ptr cloud_filtered(new pcl::PointCloud<pcl::PointXYZ>); 
sor.setInputCloud(cloud); 
sor.filter(*cloud_filtered); 
viewer->addPointCloud(cloud_filtered,"cloud"); 

Answer 1

至少，我发现在我的pcd文件中有很多点的坐标是[0,0,0]。这就是问题所在。我试图通过传递过滤器来提醒他们。而且，这个方案令人兴奋。但我想要更好地提出这些观点。

Answer 2

我用这个函数从点云中删除有限和零点，它是基于PCL Conditional removal类和isFinite()：

PointCloudPtr null_filter(const PointCloudPtr cloud_in) 
{ 
    PointCloudPtr cloud_out(new PointCloud); 
    PointCloudPtr temp(new PointCloud); 

    // build the condition 
    pcl::ConditionOr<PointT>::Ptr range_cond(new pcl::ConditionOr<PointT>()); 
    range_cond->addComparison(pcl::FieldComparison<PointT>::ConstPtr(new pcl::FieldComparison<PointT>("x", pcl::ComparisonOps::GT, 0.0))); 
    range_cond->addComparison(pcl::FieldComparison<PointT>::ConstPtr(new pcl::FieldComparison<PointT>("x", pcl::ComparisonOps::LT, 0.0))); 

    range_cond->addComparison(pcl::FieldComparison<PointT>::ConstPtr(new pcl::FieldComparison<PointT>("y", pcl::ComparisonOps::GT, 0.0))); 
    range_cond->addComparison(pcl::FieldComparison<PointT>::ConstPtr(new pcl::FieldComparison<PointT>("y", pcl::ComparisonOps::LT, 0.0))); 

    range_cond->addComparison(pcl::FieldComparison<PointT>::ConstPtr(new pcl::FieldComparison<PointT>("z", pcl::ComparisonOps::GT, 0.0))); 
    range_cond->addComparison(pcl::FieldComparison<PointT>::ConstPtr(new pcl::FieldComparison<PointT>("z", pcl::ComparisonOps::LT, 0.0))); 

    // build the filter 
    pcl::ConditionalRemoval<PointT> condrem; 
    condrem.setCondition(range_cond); 
    condrem.setInputCloud(cloud_in); 

    // apply filter 
    condrem.filter(*cloud_out); 

    pcl::PointCloud<PointT>::iterator del = cloud_out->points.begin(); 
    for (int i = 0; i < cloud_out->points.size(); i++) 
    { 
     if (!pcl::isFinite<PointT>(cloud_out->points[i])) 
     { 
      //  PCL_WARN("normals[%d] is not finite\n", i); 
      cloud_out->points.erase(del + i); 
     } 
    } 
    std::vector<int> indices; 
    pcl::removeNaNFromPointCloud(*cloud_out, *temp, indices); 

    return temp; 
}