Recent Posts

61

Python and Java API / Re: HeightField in buildModel

« Last post by Alexey Pasumansky on September 21, 2023, 06:15:50 PM »

Hello boulder1998,

Height Field mesh generation method accepts only point cloud (dense cloud) or tie points as input data and cannot work based on depth maps source.

62

Bug Reports / Re: Metashape 2.0.2 and 2.0.3 crash when using scripts

« Last post by Alexey Pasumansky on September 21, 2023, 06:11:05 PM »

Hello Wizyza,

Does the application crash, if you follow the steps using GUI only?

63

Bug Reports / Re: Assertion "3478972301441" failed at line 720

« Last post by Alexey Pasumansky on September 21, 2023, 06:02:40 PM »

Hello Patribus,

Can you please specify the NVIDIA graphic card and driver that you are using and share the log from 2.0 version related to the failed operation?

64

Agisoft Cloud / Re: Agisoft Cloud Release Notes

« Last post by Ilya Shevelev on September 21, 2023, 03:46:09 PM »

Released on 2023-09-21


Improvements

• Extended coordinate and unit systems support. Volume and Profile measurements are now displayed with the respect to selected coordinate and unit systems.

Full changelog is available in our knowledge base.

65

General / Re: Performance Recommendation for GPU Upgrade or a full make-over neccesary?

« Last post by noodlez on September 21, 2023, 02:19:59 PM »

@DayGeckoArt and @Bzuco
I must have missed to check to get notified upon replies to my post and only now came back here randomly

Thanks @DayGeckoArt for pointing out CUDA Cores - gave that a priority for the new GPU.
Also thank you @Bzuco for that hint to assess the workload of the GPU. I checked that and it seems to be more like your graph indicating that a GPU upgrade could increase performance.

So again, thank you for your help.

66

Python and Java API / Improving Orthomosaic Quality for Road Survey

« Last post by Dhanraj Jain on September 21, 2023, 02:03:39 PM »

I'm currently processing images from a road survey to generate orthomosaic and elevation TIF images, and the process is completed successfully without any errors. However, since these images are all along a single line, reflecting the nature of a road survey, I've noticed that in some parts of the orthomosaic, there are disturbances.

I'm using Agisoft Metashape Python 1.8.5 for this task, and I'm wondering if there is a special method or technique within this software that can help me address this issue and improve the quality of the orthomosaic along the road survey path. Any insights or suggestions would be greatly appreciated.

67

Python and Java API / Re: Generating DTM Images Alongside Orthomosaic, DEM, and 3D Models

« Last post by Dhanraj Jain on September 21, 2023, 01:48:52 PM »

Hello Paulo,
Thank you for your response. Could you please share a code example for this process? I'm not sure about storing DEM and DTM together or if we need to create a separate chunk for the DTM from the main chunk.
Additionally, I came across the "classifyGroundPoints" function in the Agisoft Metashape Python 1.8.5 documentation; is this the function you are referring to?

68

Python and Java API / HeightField in buildModel

« Last post by boulder1998 on September 21, 2023, 12:48:10 PM »

Dear all,

I want to play around with some of the input parameters for certain Metashape functions such as buildModel.
In my script after matchPhotos, alignPhotos and buildDepthMaps I want to use this buildModel function.

While my scripts runs fine when using Metashape.DepthMapsData as source_data, the script does not work with Metshape.PointCloudData[/b. This seems reasonable since no Pointcloud has been created beforehand (I do this later in my script).

The same approach I use for the surface_type. While Metashape.Arbitrary works, it does not work with Metashape.HeightField. Unfortunately, I could not find any further details what is behind height field.

My question therefore is: What does HeightField mean and most importantly depend on in Python?

Cheers, boulder1998

69

Agisoft Cloud / Re: Selectively upload outputs into agisoft cloud

« Last post by SFL_Beda on September 21, 2023, 11:47:30 AM »

Hi Paul

thx for the suggestion, that is certainly a way to work around this.

Here is what I tried:

Code: [Select]

    #code that creates outputs in doc.chunk and exports it locally
    doc.save()
    doclight=doc.copy()
    chunklight=doclight.chunk
    chunklight.remove(chunklight.tie_points)
    chunklight.remove(chunklight.depth_maps)
    chunklight.remove(chunklight.cameras)
    chunklight.remove(chunklight.point_clouds)
    chunklight.remove(chunklight.elevations)
    chunklight.remove(chunklight.models)
    client = Metashape.CloudClient()
    client.username = '<my_username>'
    client.password="<my_pw>
    client.uploadProject(doclight)

however I run into multiple problems:

1. when trying to copy doc to doclight, i get "Exception: Null tiled model" eveb though I can open the saved project in metashape and there is a tiled model also when looking at doc.chunk.tiled_model it clearly exists.
2.If I just apply the above directly to doc, everything but removing tie_points works. Tie_points seem to be treated differently from the rest, how do I delete them from the project?
3. After uploading (with the tie_points), the project shows up in my cloud account but looks completely empty (see attachment). I can solve that by clicking publish in the cloud but would like to do this from python. Is there a way I can enable publishing after upload (like there is the checkbox in the gui)?

70

General / Re: Need explanation,Error in pics and gcp depending of what i choose for optimizing

« Last post by Mr.Vain on September 21, 2023, 07:07:45 AM »

Hello Mr.Vain,

definitely I would say there is an accuracy offset between your GCPs and camera positions. From the accuracy, it seems the camera positions were RTK fixed (cm level accuracy). Where they determined from same base as GCPs? And maybe you can can show the error tab in reference pane to see how errors are distributed...

Thanks for your answer. Here is a screendump with error showing and the project is optimized without using the pics.