Topics

1

General / Noisier DEM from Metashape 1.7 compared with 1.6

« on: August 21, 2021, 08:28:49 PM »

Hello -- with all the same processing parameters, I find that Metashape 1.7.0 produces noisier DEMs than Metashape 1.6.5. Is this behavior known? What parameters should I change in 1.7 to get a smoother, less noisy DEM, like the ones I am used to getting from 1.6.5? See an example forest DEM from the two versions attached. Thank you!

2

General / High-resolution photos unnecessary if only using medium quality for processing?

« on: April 18, 2021, 07:45:22 PM »

Hello -- in a previous thread, Alexey explained that selecting 'medium quality' during photo alignment and dense cloud generation equates to making the photos lower-resolution, by the following factors:

For matching accuracy the correspondence should be the following:
Highest = 0
High = 1
Medium = 2
Low = 4
Lowest = 8

For depth maps quality:
Ultra = 1
High = 2
Medium = 4
Low = 8
Lowest = 16

Basically, the value defines the downscaling factor by each side applied to original image. The only exception is Highest matching accuracy, where the images are upscaled by two times by each side.

So am am wondering, if I use 'medium quality' for both steps, is there any point in capturing my photos with such high resolution? Is the original-resolution photo used for any purpose in the point cloud workflow? Or could I just capture my photos at a lower resolution (say 1/2 the resolution in x and y dimensions) and then use the 'high quality' setting instead?

Thank you in advance!

3

General / Using GCPs causes alignment problems

« on: March 07, 2021, 04:37:18 AM »

Hello -- I have noticed that in order for my GCPs to meaningfully improve my project's georeferencing, I need to uncheck all the cameras before alignment. However, when I do this, I sometimes get serious alignment problems, like in the screenshot attached. If I leave the cameras checked (default), the alignment is much better, but the georeferencing for the whole project becomes much worse. Is there something I can do differently to get good alignment and good georeferencing? My images are from a DJI Phantom 4.

Thanks in advance!

4

General / Identifying the location of a dense cloud point within a raw photo

« on: March 26, 2020, 06:38:29 AM »

Hello -- if I select a specific point in my dense cloud (or any arbitrary point in 3D space), is there any way I might be able to "back-calculate" where that point would be in one of my raw photos (i.e., which pixel of the raw photo is closest to it)? I realize this is a very unusual request, but I wanted to ask just in case someone has a clever solution.

The reason I ask is that I would like to be able to mark the top and bottom of a tree in my point cloud (or simply specifying the 3D coordinates) and then identify where those two points would be (pixel coordinates) in one of my photos.

Thanks for any insight!

5

Python and Java API / How to get a BBox from a DenseCloud?

« on: January 18, 2020, 09:53:59 PM »

Hello -- I would like to constrain the extent of an orthomosaic that is exported by the exportRaster method. I see the method has a region parameter, which takes a BBox object. I would like to constrain the exported extent of the orthomosaic to be the same as the extent of my dense point cloud. How can I get a BBox object that contains the extent (region) of my dense cloud? I checked the documentation, but there is little information on BBox objects.

Thank you!

6

Python and Java API / How to export BigTIFF orthomosaic from v1.6 Python API?

« on: January 04, 2020, 05:48:09 PM »

Hello -- version 1.6 dropped exportOrthomosaic() and now uses exportRaster() with source_data=Metashape.OrthomosaicData. However, exportRaster() does not have a tiff_big parameter like exportOrthomosaic() did. How do I export a BigTIFF using this function? I have tried the following, but it does not work.

Metashape.ImageCompression.tiff_big = True
doc.chunk.exportRaster(path="path/to/file.tif", source_data=Metashape.OrthomosaicData)

I need to use a BigTIFF because with normal TIFF settings, I get: libtiff error: Maximum TIFF file size exceeded

Thank in advance for any help!
Derek

7

Bug Reports / Error loading reflectance panel calibration CSV with Python moudle on Linux

« on: January 04, 2020, 12:47:49 AM »

EDIT: Never mind. This error resulted from a different step that I ran incorrectly.

I can't find a way to delete this message.

8

Bug Reports / 1.6 on Linux - Python module fails to save a project containing an orthomosaic

« on: January 03, 2020, 07:27:03 AM »

Hello -- if I load a pre-existing project (.psx) containing an orthomosaic using the Python API on Linux, then when I save the project under a new name using the API, I get an error.:

OSError: Can't create file: Is a directory (21): /my/path/my_project.files/0/0/orthomosaic/

This error only occurs when the project contains an orthomosaic. It works well otherwise.

9

Bug Reports / Relabeling cameras via python leads to different (worse) alignment

« on: December 05, 2019, 08:06:54 AM »

In my workflow, I add photos that are in multiple folders. Sometimes a photo with a specific name (e.g. IMG_1001.JPG) can appear in more than one folder, even though it's actually a different photo. Metashape's default label for the cameras is just the photo filename, which means I can end up with multiple cameras with the same label. This is not acceptable, because my GCP workflow requires referencing specific cameras by their labels. Therefore, I relabel my cameras to have unique labels (the label includes the name of the folder it was in):

    for camera in doc.chunk.cameras:
        path = camera.photo.path
        path_parts = path.split("/")[-2:]
        newlabel = "/".join(path_parts)
        camera.label = newlabel

However, I have noticed that if I relabel the cameras, I get a different alignment than if I don't relabel them. The alignment is much worse if I have relabeled the cameras first. For example, Metashape thinks some groups of cameras are pointing toward the sky or at very oblique angles, even though they are actually all pointing directly downward. The only thing I change between the two runs is whether or not the cameras get relabeled. I do the relabeling immediately after adding the photos. I have repeated this multiple times and I always get the same results.