Forum

Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - jmaeding

Pages: [1] 2
1
Hi there,
We fly many drone flights and compile, and there is always a warped zone around the edges where photo overlap degrades and so on.
I have run the survey statistics report tool and see some interesting things from it, but can I select pts using that?
The question posed to me was "can metashape automatically crop the pt cloud and ortho to some quality boundary?"
Of course, like HHGTTG, what does that question really mean?
I don't know, it was just asked by this guy, you know...lol (book reference)
All out of humor, send any good ideas my way if you got em.

2
General / Re: Dense point cloud to Autocad
« on: July 22, 2023, 01:38:47 AM »
"i don't know in Autocad 2022 (not Civil version) which tools to cut the cloud and create sections and terrain profiles."
Late reply but typically you create TINs from ground points and use that to do civil design with.
If you are trying to do terrain profiles without civil3d, look into Carlson or Dotsoft tools for autocad.
Not free, but $2000 in tools is only 20 hours if billing $100 and hour. You will waste far more than that without decent civil tools.

3
Oh, very good tip on setting the camera z accuracy low to let GCP dominate Z.
Thanks again for the reply!

4
Dieter,
I use TBC, which you may have a seat of if you have realworks.
TBC can export .rcp format, and makes them 10x faster than recap.
You do have to have recap installed, but do not need a license for it, free is fine.

I was jumping for joy when I figured this out, recap is something I avoid these days.

5
I've never heard of that working. The problem is the software is not able to match up enough points to line up the added images.
I'm guessing the terrestrial pics are from hilltops or sides of buildings?
Even shadows that are different than time of drone flight will mess that up.
The terrestrial camera has no coordinates/direction info, right?
So projecting stuff won't help.
I like the idea though, and if you took pics on ground at flight time, that would be useful to add.
Seems like you would have to take multiple overlapping ones though.
This falls into the category of feeding metashape drone and ground photos to get one model.
Anyone done that without garbaging the results?


6
General / Re: PC for Metashape in 2023
« on: June 08, 2023, 02:14:46 AM »
My experience is the extra 8 cores are definitely worth it.
64 gb ram should be plenty.
I have an nvidia card with 4 gb ram like you and its very good.
I would recommend a secondary internal SSD drive of 1 tb.
Run your projects on that, and then dump to external drive to make space later.
1500 photos is a lot, I typically do 400 or less for my drone flights.
The 20 laser scans part seems huge to me.
Maybe if doing mechanical stuff they are not too bad, our civil ones are 100 million a scan.

The other thing is its not great to buy high end now, and have it last 5 years.
Its better to go medium high now, and buy new in 3 years. Things change too fast.

7
@Paulo,
Ah, very useful, thanks.
I also posed the question to support so getting good info from both sides.
There is another aspect to this that actually may be more important:
Does metashape mix together the photo coords and GCP's and come up with one set of control it uses?
Or does it form a "photo network" from photo info, then apply gcp correction to that?

What is going on with me is I have ppk data that seems perfect horizontally, but the vertical seems off by about a foot.
If I then add in 4 gcp's and run, with both photos and markers checked, the results seem on vertically.
But, how are areas away from the markers affected? Do they try to stick to location based on photo coords?

Metashape behavior is very clear to me if just using photo coords, or just gcp's.
Its when they are mixed that I have to be careful about fooling myself.
thanks

8
Hello,
We have a DJI Mavic 3E, and use a ppk workflow. Surveyors run a base station but we don't hook to it (no RTK).
We process our photo locations through TBC and get a text file with adjusted camera locations for metashape.
We use that text file as reference and works well.
TBC has an issue where the elevations are lifted up for some reason by 3 or 4 ft.
We typically set 4 GCP's which we have correct coords for (no error like ppk data...), and when we add those in, the elevations are corrected.
Why is that though? We have 300 cameras and only 4 gcps.

The question is how the GCP's modify the results, compared to just camera locations used.

This is quite important to know, but also likely has a few answers.
Is there anything "clear" than can be said about adding gcp's, like "they dominate the vertical placement of the results"?
This topic is super important to understand when discussing why gcp's are still recommended in ppk workflows.
In our case, our ppk elevation values have issues which we will fix eventually.

thanks,

9
I've been using metashape for a few years now (was PS), and ran into somthing I need to understand.
I'm a civil that flies a phantom 4 pro, and has GCP points on the ground.
I use the USGS workflow I have posted here before, and generally get super good clean results.
I had a run though, where after processing the control points show quite a bit above the ground.
It occurred to me to uncheck the markers and I realized there is a "real" location, and "compiled" one - wow.
If I uncheck the markers, they show at coords I gave them and are right on the ground.
Now I am wondering what to think of this. Why would the results be essentially correct, yet the markers moved vertically after processing?
I know unchecking makes them check points. Does a large difference (more than 1 foot) between original and final location indicate anything is bad?

I'm literally facing my boss and client, wondering if I should use the data or discard and refly.
thanks

10
Hi Alexy,
Thank you so much for replying, this was eating away at me.
I figured out what you mean, uncheck the points so they behave as check points, then recheck once refined.
That does work, just tried on latest version.
I have to say, I think the change you made that causes this was a very detrimental to metashape's adoption.
The workaround is rediculously easy, no complaints whatsoever there.
What I mean is people new to the tool will never figure that out without a lot of trouble.
They will conclude the marker alignment is a time waster, and go to another software.
That is a horrible consequence of just a minor default setting change.
I strongly suggest you set the default back.
Maybe I missed some way to do that myself, but all the people flying drones for mapping will run into this.
Maybe the other user types will not get bit by this, but its important enough you should do a nag screen asking what type of project it is.
I just don't want such a great tool at the best price by far, to be swept aside by potential customers.
Keep up the good work!

11
There is still something not working like it used to. The markers are not adjusting after the first few are placed.
I need to contact support as this seriously slows down my workflow.
I would go back to whatever version this worked correctly in no matter how far.
I'll get with support.

12
So I tried running optimize cameras after adding the control points, but not adjusting them.
I never did this before, but it gets the tie points much closer, then things seem to work better.
I still think something changed, but simply adding the optimize step (all General boxes checked) after adding control points is enough.
This makes me think the optimize options are affecting things. Like that additional corrections item.
Tricky to know what all that does as its not documented that I could see.

13
I have used PS, now metashape, for a few years and same drone (P4P V2), and always follow the same workflow.
I use the USGS workflow so am doing things super common to most.
Lately, maybe in last 2 months, when I pull in text file with ground control points, the markers come in fine.
Then I go through my regular steps of:
1) right click in model by a marker, then "filter photos by point". that works.
2) place or drag the markers in a couple photos to refine the location, turns them to green flags.
3) NORMALLY - that causes the same marker in other photos to move very close to those first two.

Now, the markers are not adjusting, and I have to move them a large amount in the remaining photos.
Many times I have to hand place them, they don't even show.

This is much slower than before, as I can't even use the "filter photos by markers" once I adjust a couple, as that is better than filter by point.
I have been trying to keep up with latest version but not sure when exactly this started happening.
I have no issues otherwise. GPS data on photos is fine, aligning cameras is fine...its all good.

What settings could affect the other markers to keep them from self adjusting closer to the first couple I hand adjust?
thanks

14
Hmm, I did not have the log file set.
Funny thing though, I originally cancelled the ortho creation after seeing it would take 2 hours.
Then I realized trying to use PS might not work on a project opened and saved with metashape, so I reran the ortho creation in MS.
The second time, it ran in normal 20 minute time frame.
I'm going to test more to look for this pattern, and also have turned on the logging.
thx

15
welp, got no replies, but I think the answer is no. The point cloud will not match the GCP's exactly, there is some kind of fit going on.
As I was investigating this on a particular job, I read how you can set mode to "estimated" and see what the software got for each control point after processing.
Then I realized I should be paying attention to the error values before processing, as those will indicate the difference between the GCP numbers and where the model thinks they should be (after registering markers on the photos).
In the end, the lesson I learned is if you get error values higher than 1 ft, there is a mistake in the GCP coordinates, or the photo registration.
One typical mistake is duplicate elevation values in the GCP's from copying and pasting numbers wrong.
Another is placing markers on a spot that is not the same for a given GCP.
I added a step (19) to my essentially USGS workflow, which is summarized here:
Condensed Workflow

1)   Make project and load photos, leave checked
2)   Convert to State plane, ESPG::2230 = Zone 6, ESPG::2229 = Zone 5
3)   Check that GPUs are set to be used in Tools->Prefs->GPU tab
4)   Align Photos, accuracy on high
5)   Show projection and error columns in photo list
6)   Do optimize (not all boxes checked)
7)   Do Gradual Selection using Reconstruction Uncertainty, select 20% at a time, goal is 10. Repeat 6 and 7 until you get close.
8)   Do optimize (not all boxes checked)
9)   Do Gradual Selection using Projection Accuracy, select 20% at a time, goal is 2.5. Repeat 6 and 7 until you get close.
10)   Open reference settings and set Tie Point Accuracy to 0.1, Do Optimize (not all boxes checked)
11)   Uncheck photos as we are done with using their gps tag data
12)   Create Control Points list as csv
13)   Import csv to create control point markers, say Yes to All when asked
14)   Right click on a point by marker, and choose Filter photos by point.
15)   Move marker to correct location in 2 photos. This refines the locations.
16)   Repeat right click on a marker, and choose Filter photos by marker this time.
17)   Go through each photo, and if target is clear in the photo, drag the marker to the correct location. Grey flag means “not used”. Green flag means “used”. Hit page down to go to next photo. It should be click, page down, click, page down…
18)   Block any markers, using right click, that you decide are not clear, if already set green.
19)   Double check that none of the markers have high error values. Also look for duplicate elevations indicating the elev was copied from another spot on accident.
20)   Do optimize (all boxes checked)
21)   Do Gradual Selection using Reprojection Error, select 10% at a time, goal is 0.4. Repeat 6 and 7 until you get close.
22)   Build dense point cloud medium level
23)   Build DEM (needed for ortho)
24)   Build Ortho photo (machine)
25)   Export Laz point cloud file
26)   Export Ortho photo as jpg with world file. Set compression to jpg and select jpg in file type in dialog.
27)   Create .rcp file in Recap for Autocad/Infraworks use.

Pages: [1] 2