Trimble Business Center

I am on my 3rd or 4th photogrammetry project and I am noticing a trend.  In the Adjustment report the residuals are pretty good, the one I am looking at now has a max vertical of 0.109' and a mean of -0.0282', really awesome.  When I take the same points and compare them with the surface made with the LAS file they are always off, max is 0.489' and the mean is 0.20'.  I thought it may be the way I am creating the surface so I inversed between the GCP and the nearest point and it is really close to the point to surface report.

Do I have a setting off or something?

Equipment: Mavic 3E, R12 static, PPK, 150 AGL, 6 GCP, and 12 check points, TBC 24.10.9090, roughly 40 acre, 80% overlap

Sam

This is normal. When you are producing a surface, presumably you are downsampling the point spacing. Likely via the advanced filtering tool. in my experience, the points to surface report will always be quite a bit coarser than simply comparing points in the point cloud against control/check shots. be sure to clean the surface to remove any erroneous peaks or valleys, and of course, use breaklines.

I am on my 3rd or 4th photogrammetry project and I am noticing a trend.  In the Adjustment report the residuals are pretty good, the one I am looking at now has a max vertical of 0.109' and a mean of -0.0282', really awesome.  When I take the same points and compare them with the surface made with the LAS file they are always off, max is 0.489' and the mean is 0.20'.  I thought it may be the way I am creating the surface so I inversed between the GCP and the nearest point and it is really close to the point to surface report.

Do I have a setting off or something?

Equipment: Mavic 3E, R12 static, PPK, 150 AGL, 6 GCP, and 12 check points, TBC 24.10.9090, roughly 40 acre, 80% overlap

Sam

Let me make sure I am on the same page, when inversing from the nearest point cloud point to point 201 (a target check point) I am getting 0.489ft, no surface involved.  In the photogrammetry adjustment report 201 has a vertical residual of -0.07.  Also I will add that all the photogrammetry projects have this same problem, and it is always in the same direction.  The target points are below the point cloud points.  I could see if they varied above and below, with a mean of around zero, but they are always lower than the cloud and have an average of about 0.25 ft.  This seems to be a systematic error.

Sam

This is normal. When you are producing a surface, presumably you are downsampling the point spacing. Likely via the advanced filtering tool. in my experience, the points to surface report will always be quite a bit coarser than simply comparing points in the point cloud against control/check shots. be sure to clean the surface to remove any erroneous peaks or valleys, and of course, use breaklines.

I am on my 3rd or 4th photogrammetry project and I am noticing a trend.  In the Adjustment report the residuals are pretty good, the one I am looking at now has a max vertical of 0.109' and a mean of -0.0282', really awesome.  When I take the same points and compare them with the surface made with the LAS file they are always off, max is 0.489' and the mean is 0.20'.  I thought it may be the way I am creating the surface so I inversed between the GCP and the nearest point and it is really close to the point to surface report.

Do I have a setting off or something?

Equipment: Mavic 3E, R12 static, PPK, 150 AGL, 6 GCP, and 12 check points, TBC 24.10.9090, roughly 40 acre, 80% overlap

Sam

Hmm. I didn't know that you could inverse between a survey point and point cloud point in TBC. Is it properly snapping to the survey point AND point cloud point? 

Let me make sure I am on the same page, when inversing from the nearest point cloud point to point 201 (a target check point) I am getting 0.489ft, no surface involved.  In the photogrammetry adjustment report 201 has a vertical residual of -0.07.  Also I will add that all the photogrammetry projects have this same problem, and it is always in the same direction.  The target points are below the point cloud points.  I could see if they varied above and below, with a mean of around zero, but they are always lower than the cloud and have an average of about 0.25 ft.  This seems to be a systematic error.

Sam

This is normal. When you are producing a surface, presumably you are downsampling the point spacing. Likely via the advanced filtering tool. in my experience, the points to surface report will always be quite a bit coarser than simply comparing points in the point cloud against control/check shots. be sure to clean the surface to remove any erroneous peaks or valleys, and of course, use breaklines.

I am on my 3rd or 4th photogrammetry project and I am noticing a trend.  In the Adjustment report the residuals are pretty good, the one I am looking at now has a max vertical of 0.109' and a mean of -0.0282', really awesome.  When I take the same points and compare them with the surface made with the LAS file they are always off, max is 0.489' and the mean is 0.20'.  I thought it may be the way I am creating the surface so I inversed between the GCP and the nearest point and it is really close to the point to surface report.

Do I have a setting off or something?

Equipment: Mavic 3E, R12 static, PPK, 150 AGL, 6 GCP, and 12 check points, TBC 24.10.9090, roughly 40 acre, 80% overlap

Sam