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 - JMR

Pages: 1 ... 3 4 [5] 6 7 ... 30
General / Re: Number of Keypoints
« on: August 31, 2020, 02:49:31 PM »
Yes. you can dig into project folder structure and find a file called doc.xml inside of the latest folder located in project.files/0/0
There is a slight problem that it is to identify image_id with its label or file name. You need to open another one, equally called doc.xml packed inside the folder, located in project.files/0/

There must be a simpler way by means of python console, but I can't help with that.

General / Re: Extracting camera trigger times
« on: August 25, 2020, 10:16:20 PM »
If time stamps have been written to photo exif by some on-purpose hardware solution (like in a Phantom 4RTK) then you can get up to millisecond time resolution and likely you can batch export them with exiftool. But if you just pretend to use the standard time stamp of your camera for precisely calculate camera location... I'm afraid you are not in the right way. Time tag is written with one second resolution in most cameras. If the camera is moving at 5m/s (quite a normal speed for a drone), your interpolation in camera path will lead to higher uncertainty than your gps.
Most systems created for getting high precision geotags, write time-stamps of shot events into a log file, not in the photo metadata. I guess this is the file you need.

General / Re: Appemd Question
« on: August 25, 2020, 10:00:47 PM »
It does a summation of projects by adding another project as a new chunk of the current one. If the project that is to be appended contains several chunks, the user will be asked if wants to add all or only some of them. Image paths will remain unchanged, while the current project folder will receive a copy of the contents of the appended project.
The appended project will remain unchanged.
Try and learn yourself.
Best regards

General / Re: clip to boundary shapes
« on: August 24, 2020, 10:30:24 AM »
Hola Javier. IN order to have this option enabled, you need first to tell the program that this shape is a boundary, either outer or inner. To do so, select it, right click and set boundary type.
Best regards;

General / Re: Can't add additional photos for alignment
« on: August 24, 2020, 10:27:46 AM »
Have you checked the keep tie points option? if not, do it before first run of align command. That should fix your problem.
Best regards

General / Re: GCP error margin 20-30 cm
« on: August 22, 2020, 01:59:43 PM »
Some comments to your details.
You should not use highest accuracy with your camera. High uses full image size and is actually the highest recommended setting for alignment (from the user manual "Highest accuracy setting is recommended only for very sharp image data and mostly for research purposes")
FZ300 is a good camera for many purposes, but you need to turn off the OIS otherwise it can ruin your photogrammetry project. Also autofocus should be disabled and focus set manually once according to the shooting distance for best results.

Estimating tie points covariance would give you chance for filtering and blunder detection. Also by analyzing the pixel errors in the right most column of your ref table, you would be able to see if photos are badly aligned. If error are clearly over 1 pix, then it is likely the alignment is not right.

I think you should not have cleared gps info from uav images. If you want to ignore global position, do it after alignement by making null the global transform. but during alignment, geotags will be helpful. I have never seen errors like 30cm in any P4Pro dataset.


General / Re: GCP error margin 20-30 cm
« on: August 20, 2020, 10:49:11 AM »
some more about the circumstances can help to find possible causes.
Do cameras geotag images? likely in uav, but not so much in pole mounted camera...
if so... what is the camera centers accuracy in ref settings?
what is the  gcp accuracy settings?
what the overlap?
how is like the distribution of GCP with regard to the mapped area?
what about cameras pixel error after alignment?
what about tie points covariance?
Is the issue happening always with the same camera/lens? if so, have you checked the camera?
is that the same camera/lens that you have used for previous successful projects?
does it have rolling shutter? (for example, note that some Panasonic cameras, even if they have a mechanical shutter, they use electronic shutter for speeds beyond 1/500)

I hope we can help if you if you provide details

General / Re: Seamline reuse
« on: August 18, 2020, 08:14:28 PM »
Hi, Davier:
Since version 1.3.1 it is doable by means of API functions, but also, one could do it manually by using shapes for assigning source photo for each... possible but cumbersome.
I can't help with programming the script, but I'm 100% sure it is something that someone has done before.
Best regards.

Feature Requests / Re: Selection Toolset -> Add Polygon Tool
« on: August 09, 2020, 02:47:48 PM »

Feature Requests / Spherical targets
« on: August 03, 2020, 08:59:30 PM »
Hello, Agisoft:
Spherical targets are widely used in terrestrial laser scanning. They are convenient due to de fact that a sphere looks the same from any pont of view and does not require being orented towards the sensor like planar targets do.
I guess it would be really usefull if Metashape where able to detect a sphere centre point automatically as it does with circular targets. It would help laser scanning works to be co-registered with photo scans.
Unlike planar circles, spheres always look round enough in a picture regardless the angle of view, so a circular feature detector, should work fine, even when it is not surounded by a black crown like in non coded circular targets.
Please keep improving Metashape by adding funtions for a better integration with TLS.
Best regards

General / Re: Export images with applied shapes
« on: July 27, 2020, 10:28:06 PM »
Hi Jaume: why do you wnat shapes rasterized onto individual photos? oe why not onto the orthomosaic? the second is an easy to do task, but superimposing on the photos is kinda dificult. For example keep in mind that a straight line that you draw on the model needs to be distorted (applying radial distortion) to be rasterized onto photos accurately, or a line that is not lying on terrain, will show some offset when reprojected on a photo.

General / Re: P4 RTK questions workflow error in convertin?
« on: July 15, 2020, 02:37:05 PM »
Are you 100% confident that you have set up rtk reference coordinates correctly? You might do a cross check by doing PPK of your dataset and see if embeded geotags are consistent with PPK results.
In my experience Metashape is not always as good as competitors in alignment of projects without any ground control point (direct referencing) . It is sad for me to say so, but I've found it in a few cases.
If you want good results, you need to calibrate your camera in a very well controlled project and then use found lens parameters as precalibrated.

It is hard to predict and also to know dense cloud density (or average points spacing, if you like) but once it is created it is easy to filter to a desired spacing. So you could be able to subsample yor cloud to a value that is fine for you... yeah, it soulds like climbing to the top of a mountain and then descend to a desired altitude instead of climbing directly to that point, but... I'm sorry, no straight path in this case as far as I know.

General / Re: Workflow - Dense cloud generation ignoring masks
« on: July 10, 2020, 01:33:22 AM »
well, you can try now select dense points by mask and crop.
select by colour in your very case could also have good results.
Good luck

General / Re: Create contour by Shape or by Selection?
« on: June 23, 2020, 10:04:00 AM »
Boundaries are used only at export time, so what if you export the dem limited to your boundary and then re-import it on the same chunk?
I guess it would work as you want.
Kind regards

Pages: 1 ... 3 4 [5] 6 7 ... 30