Unable to produce clustering

[vincentcheungm]

Unable to produce clustering

[VincentCheungM]

Maybe round is enough.

[VincentCheungM]

Yes, it's mainly due to the process in here.

    // Organize Pointcloud in scanline
    SLRPointXYZIRL point;
    for(int i=0;i<laserCloudIn.points.size();i++){
        point.x = laserCloudIn.points[i].x;
        point.y = laserCloudIn.points[i].y;
        point.z = laserCloudIn.points[i].z;
        point.intensity = laserCloudIn.points[i].intensity;
        point.ring = laserCloudIn.points[i].ring;
        point.label = 0u;// 0 means uncluster

        if(point.ring<=sensor_model_&&point.ring>=0){
            laserCloudScans_[point.ring]->push_back(point);
        }    
    }

It's re-organizing the points within point cloud to the correspond scan line. The ring is the scan line index of the points, which can be achieved with the help of Velodyne drive. However, the binary point cloud files from KITTI don't provide the ring property, and it goes wrong. So you can calculate the ring by your self or using other point cloud files.
The ring can be calculated in this formula(latex code):

$$ring =\arcsin{\frac{z}{\sqrt{x^2+y^2+z^2}}}$ $

[lykhahaha]

Thank you for your help.I also think the reason for the failure of clustering is because of the lack of rings.
The result of this formula is angle(float), but the value of rings should be 0 to 63 (int). How should I convert?