Messages

1

Bug Reports / Re: Export Stereo Files Error

« on: August 20, 2024, 09:30:45 AM »

All cameras are aligned.
Am running 2.0.1 exe to get correct outputs, not and ideal solution.

2

Bug Reports / Re: Export Stereo Files Error

« on: August 08, 2024, 02:13:36 AM »

Hi Alexey,

No single camera system  Multiple images

Below is a sample of the stereo files that used to be exported when selecting Inpho Project File (*.prj)

$PROJECT 5.0.0
  $STARTING_DATE : Sun May  1 12:10:19 2022
  $LAST_CHANGE : Sun May  1 12:10:19 2022
  $COORDINATE_SYSTEM :
     PROJCS["GDA94 / MGA zone 55",GEOGCS["GDA94",DATUM["Geocentric Datum of Australia 1994",SPHEROID["GRS 1980",6378137,298.257222101,AUTHORITY["EPSG","7019"]],TOWGS84[0.06155,-0.01087,-0.04019,0.0394924,0.0327221,0.0328979,-0.009993999999999999],AUTHORITY["EPSG","6283"]],PRIMEM["Greenwich",0,AUTHORITY["EPSG","8901"]],UNIT["degree",0.01745329251994328,AUTHORITY["EPSG","9102"]],AUTHORITY["EPSG","4283"]],PROJECTION["Transverse_Mercator",AUTHORITY["EPSG","9807"]],PARAMETER["latitude_of_origin",0],PARAMETER["central_meridian",147],PARAMETER["scale_factor",0.9996],PARAMETER["false_easting",500000],PARAMETER["false_northing",10000000],UNIT["metre",1,AUTHORITY["EPSG","9001"]],AUTHORITY["EPSG","28355"]]
  $LINEAR_UNITS_OF_OBJECT : m
  $LINEAR_UNITS_OF_IMAGE : mm
  $ANGULAR_UNITS : deg
$END
$AAT
$END
$PHOTO
  $PHOTO_NUM : TIP11_2021-22_R1_0001
  $PHOTO_FILE : D:/TIP11/Processed/TIP11_2021-22_R1_0001.jpg
  $CAMERA_ID : iXM-RS150F,_Rodenstock_RS-90mm/Aerial_(90mm)
  $TERRAIN_HEIGHT : 425.8832780679744019
  $PIXEL_SIZE : 0.0037600000578394
  $SCAN_POS : 0
  $IO_PARS : 2 2
  $IO_STAT : manual
  $EXT_ORI : 12:10:19 01/05/2022
        89.4626678100099895    509549.7793335155001841   5237518.5711281569674611       3062.7218916174288097
      0.9964456707226194      -0.0818030219443087      -0.0201069863213849
      0.0819291782884282      0.9966228044823754      0.0055313046081774
      0.0195866036650639      -0.0071589933974339      0.9997825332393058
$END
$PHOTO
  $PHOTO_NUM : TIP11_2021-22_R1_0002
  $PHOTO_FILE : D:/TIP11/Processed/TIP11_2021-22_R1_0002.jpg
  $CAMERA_ID : iXM-RS150F,_Rodenstock_RS-90mm/Aerial_(90mm)
  $TERRAIN_HEIGHT : 415.4315673365714474
  $PIXEL_SIZE : 0.0037600000578394
  $SCAN_POS : 0
  $IO_PARS : 2 2
  $IO_STAT : manual
  $EXT_ORI : 12:10:19 01/05/2022
        89.4626678100099895    509582.4645049434620887   5237742.1839126134291291       3062.1639857569516607
      0.9966167386214771      -0.0816266279943974      -0.0096005157025179
      0.0817885490283871      0.9964890469962620      0.0178944813926767
      0.0081061425672820      -0.0186191519341254      0.9997937875551803

Now they look like this below with only a single photos exterior orientation outputted and a whole series of Photo Points
Something is not working as it should

$PROJECT 5.0.0
# generated using Agisoft Metashape version 2.1.0 build 17532
  $STARTING_DATE : Thu Aug  8 09:08:28 2024
  $LAST_CHANGE : Thu Aug  8 09:08:28 2024
  $COORDINATE_SYSTEM :
     PROJCS["GDA94 / MGA zone 55",GEOGCS["GDA94",DATUM["Geocentric Datum of Australia 1994",SPHEROID["GRS 1980",6378137,298.257222101,AUTHORITY["EPSG","7019"]],TOWGS84[0.06155,-0.01087,-0.04019,0.0394924,0.0327221,0.0328979,-0.009993999999999999],AUTHORITY["EPSG","6283"]],PRIMEM["Greenwich",0,AUTHORITY["EPSG","8901"]],UNIT["degree",0.01745329251994328,AUTHORITY["EPSG","9102"]],AUTHORITY["EPSG","4283"]],PROJECTION["Transverse_Mercator",AUTHORITY["EPSG","9807"]],PARAMETER["latitude_of_origin",0],PARAMETER["central_meridian",147],PARAMETER["scale_factor",0.9996],PARAMETER["false_easting",500000],PARAMETER["false_northing",10000000],UNIT["metre",1,AUTHORITY["EPSG","9001"]],AUTHORITY["EPSG","28355"]]
  $LINEAR_UNITS_OF_OBJECT : m
  $LINEAR_UNITS_OF_IMAGE : mm
  $ANGULAR_UNITS : deg
$END
$AAT
$END
$PHOTO
  $PHOTO_NUM : TIP11_2021-22_R1_0001
  $PHOTO_FILE : F:/TIP11/Processed/TIP11_2021-22_R1_0001.jpg
  $CAMERA_ID : iXM-RS150F,_Rodenstock_RS-90mm/Aerial_(90mm)
  $TERRAIN_HEIGHT : 425.883278
  $PIXEL_SIZE : 0.003760
  $SCAN_POS : 0
  $IO_PARS : 2 2
  $IO_STAT : manual
  $EXT_ORI : 09:08:28 08/08/2024
        89.462667810010    509549.779333515500   5237518.571128156967       3062.721891617429
       0.996445670723      -0.081803021944      -0.020106986321
       0.081929178288       0.996622804482       0.005531304608
       0.019586603665      -0.007158993397       0.999782533239
  $PHOTO_POINTS :
    100000000          513.60242        43.15606   1.00  3 { * }
    100000001         1382.95020       101.12618   1.00  3 { * }
    100000002         1961.14453        39.08981   1.00  3 { * }
    100000003         1941.60095        74.24004   1.00  3 { * }
    100000004         1985.79346        58.25944   1.00  3 { * }

3

Bug Reports / Export Stereo Files Error

« on: August 07, 2024, 09:06:33 AM »

Export of Camera Inpho Project File  .prj is giving strange results.
Only generates a file for a single image instead of exterior orientations for each frame in the project, and the file it produces is wrong.
Please fix it.

4

Bug Reports / Re: Is the Mosaic blending in texture generation broken with GPU acceleration?

« on: March 15, 2023, 06:10:38 AM »

I am still seeing substantial performance issues with Image blending in the orthomosaic build process when using averaging. I am using version 2.0.1
GPU use is non existent and CPU use is minimal. Its taking in excess of 200+ seconds for the blending of individual image groups to occur.
Also in the batch processing settings for build orthomosaic you can refine seamlines when using average for mosaic, in the workflow dialogue you cannot.  Is this a bug?
Its taking inexcess of 2 days to generate a ortho from 800+ images. Its never been this slow in previous versions.

5

General / Re: optimizing co-alignment from very different cameras/views

« on: June 28, 2022, 05:20:56 AM »

Hi Andyroo, I have been working with these issues for a while in the terrestrial historical photo matching area. Its not easy.  I am finding that each scenario (difference in resolution or gamma) generally requires a different approach. Bare in mind that i am working with a combination of 150mp digital aerial photos from a phase one and 600 DPI scans off historical prints or films at scales from 1:500 to 1:40000. Some disparate resolution sets align much better on medium and low alignment settings and some on maxed out settings. In most cases just the addition of three to four control points even if not at high precision makes a massive difference. Once you have a match or block of matches then incrementally selecting strings of images adjoining matched images and aligning them also works. Once you have a base dataset aligned you can use that as a base image to align more recent blocks too. Happy to keep om talking.

6

Bug Reports / Re: Invalid shape - polygonal

« on: February 11, 2022, 01:54:54 AM »

Thanks Paulo,

That seems to have done the trick.
Weird that the shapes work with the old version though.
My thought is that Metashape is importing the shapes as multipolygon in error.

7

Bug Reports / Re: Invalid shape - polygonal

« on: February 09, 2022, 06:24:42 AM »

Hi Alexey,

I am still having trouble with importing a shape to set as the boundary for Orthomosaic clipping.

I have attached a kml that enables you to set the type as Boundary in version 1.5 but will not give that option in version 1.8

Can you advise a workaround other than loading the project in 1.5 to do it.

Many Thanks

Matt

8

Bug Reports / Can't pack file error

« on: May 06, 2020, 03:27:39 AM »

I am getting an error i have not seen before. 
While building a 6000 50 MP image orthomosaic I am getting a Cant Pack file error (see attached image).
I gave 500 gig spare on the System Disk and 1TB spare on the data drive and @ 380 gig of RAM. 
Any help would be appreciated

----> Well it seems it was because i was using .cr2 files to build the Ortho.  They have been supported in the last few builds for point detection/alignment but when writing the orthos the images are not optimized and are too large for my 2 TB M2 disk.

9

General / Re: Change images names after Align photos

« on: April 28, 2020, 05:22:34 PM »

Hi Alexey,

Can you please update the rename photo script below for use in version 1.6.

Many Thanks

Matt

import PhotoScan

chunk = PhotoScan.app.document.chunk

string = PhotoScan.app.getString("Input prefix to be added to the selected camera labels:")
for camera in chunk.cameras:
   if camera.selected:
      camera.label = "_".join([string,camera.label])
      
print("Script finished")

10

Bug Reports / Re: Shift of ortho image upon exporting

« on: March 30, 2020, 01:11:07 PM »

Try resetting the Z axis in the model. There is a python script I have used in the past attached.
This can sometimes shift out of alignment when dealing with old imagery.

11

General / Re: Multicamera alignment issues

« on: September 17, 2019, 04:40:30 AM »

Hi Echostorm and Alexey,

I have been having similar problems trying to align images from an A3 aerial surveying camera.
I found that using Photoscan 1.2 - 1.4.3 I was able to align the images with more success. If using the default settings High 40000/4000 doesnt work I found that aligning at low using 20000 key points and 4000 tie points or loosening the marker and tie point accuracy may also help with your alignment issues. 
If you still have issues i suggest using some tie points or control points.
Alexey can you please supply some feedback as to why the more recent versions of Metashape are not aligning complex camera scenarios.
I have been using Agisoft since beta and alignment while a lot faster now is not working reliably for my needs.

Many Thanks

12

General / Re: How to loosen tolerances in alignment to deal with minimal or poor image overlap

« on: September 02, 2019, 04:03:45 AM »

Use Photoscan Version 1.4.3 and see how you go.  Otherwise manual tie points with no coordinates can work well.

13

General / Re: Metashape 1.5 build 7492 align issues

« on: September 02, 2019, 04:00:48 AM »

HI Alexey,

I am also having alignment issues.  I have had to use Photoscan 1.4.3 to get my 21 GB photo dataset to align.
I have used the same settings and am getting vastly different results. Metashape is very quick but it simply does not align.
See attached

14

General / Re: Help removing the "Bowl Effect"

« on: August 14, 2019, 03:35:01 AM »

You are better off not using your predefined camera focal length and letting photoscan work it out for you during alignment.
Mask the edges of the images and off you go.  I have successfully applied this methodology to scanned photos from the 1930's and with good ground control you can georeference the cloud and produce an orthomosaic to within a pixel or two.

15

Feature Requests / Re: Dense Cloud Classification settings in chunk info

« on: December 12, 2017, 02:09:47 AM »

Hi Alexey,

Yes, info saved to metadata information would be good (ie show info from right clicking on chunk)

It would be good if the settings in the classification panel were Photoscan Default initially (when unclassified) and then kept as the last classification settings post classification for each chunk. This would be good for me as I capture data at three different GSD scales  5cm, 2.5 cm and 1 cm and process them in individual chunks.

Thanks Alexey

The software seems to be getting better and better. Great work