Topics

1

General / Current best 360 camera?

« on: February 28, 2024, 01:18:36 PM »

I'm looking to buy a 360 degree camera, which will be used for both videos and photogrammetry. What is the current best available consumer or prosumer 360 camera please, in terms of photogrammetry? They could be up to $5000, but preferably less than $1000 (USD). It'd be used both indoors (lower light) and outdoors on high poles (where cases where drones are prohibited).

Many thanks in advance.

2

Feature Requests / Ability to input more parameters from Camera Calibration Reports/Certificates

« on: June 13, 2023, 07:06:52 PM »

Currently Metashape will accept the following camera calibration parameters (note those in bold):

f - Focal length measured in pixels (in pixels).
cx, cy - Principal point coordinates, i.e. coordinates of lens optical axis interception with sensor plane (in pixels).
b1, b2 - Affinity and non-orthogonality (skew) coefficients (in pixels).
k1, k2, k3, k4 - Radial distortion coefficients (dimensionless).
p1, p2 - Tangential distortion coefficients (dimensionless).

It is also possible to input the coordinates of any fiducial marks in mm, and also to input the focal length in mm.

Many archive (historical) aerial photographs (airphotos) come with Camera Calibration Reports/Certificates. The calibrated focal lengths and coordinates of the fiducial marks can readily be input. Where there are many airphotos in a strip or block these are usually sufficient initial conditions, which can readily be improved upon (i.e. the other parameters can be reliably estimated). However, it would still be useful to be able to add the many other calibration parameters available in certificates/reports, and especially in instances where very few photos are available (e.g. just 2 to 4) and thus the estimates for the other parameters listed above are less reliable.

One of the barriers to this is the units used. For instance, most certificates/reports list the coordinates of the principle point in mm (whereas Metashape only accepts pixels). This can be estimated manually, if one knows the fiducial coordinates in mm and the number of rows/columns between them... but it would be much better to have the additional option of adding Principal point coordinates in mm please.

In a similar way radial distortion is usually listed in the certificates/reports, but in um in relation to degrees. It would be great to have the ability to optionally add the distortion in such units please.

In some cases certificates/reports include calibrated distances between (i.e. across) pairs of fiducials (in mm). It would be great to have the option to add these also please.

Finally, some certificates/reports include uncertainties (in mm) for the estimate of the calibrated focal length. It would be great to be able to add these too, for use in constraining the SCBA.

Below is an example Camera Calibration Report/Certificate listed the parameter and units I mention.

Many thanks indeed.

3

General / Migrating data: Can't Read from Source File or Disk Error

« on: April 12, 2023, 02:41:03 PM »

I'm migrating data and products from a smaller drive to a larger drive on my Windows 10 PC (all drivers etc. up to date). I'm migrating all kinds of data, from lots of different sources, and products from lots of different image processing and GIS packages. Everything has gone well, apart from several Agisoft Metashape files, which stubbornly will not copy across due to the error "Can't Read from Source File or Disk Error".

At first I thought this was a problem with the small drive (e.g. bad sectors, mis-matched formats etc), but CHKDSK and other checks show all is well. I also restarted several times and tried copying to different drives instead. It was then that I noticed Metashape files were the ONLY files which had this problem... no other files from any other source, no matter what the size or format, suffered this issue. Further, it seems to be specific file types, such as:

images*.zip (e.g. images0.zip, images1. zip)
dense_cloud.oc3

Any help in fixing this would be much appreciated 

4

Feature Requests / Orient model using "scale bar"

« on: January 12, 2023, 12:13:49 AM »

I have an unreferenced model (no targets used during or after image capture). I've scaled the model using several "scale bars" created from pairs of markers. I've tried manually rotating/orienting the model such that it is the right way up and facing the correct direction, but this is very difficult to do accurately. I know that the marker pairs, and therefore the scale bars which join them, happen to be aligned with the axes of the object in real life (for instance at least one scale bar is parallel to X, another is parallel to Y and another to Z, in real life). Therefore is  there any means by which I can use the scale bars to orient the model of my object to the real-world X, Y and Z axes please?

If not, could this be added as a Feature Request please.

Many thanks

Andy

5

General / Using Leica DMC III images

« on: August 02, 2021, 06:23:26 PM »

I would like to use some stereo, vertical Leica DMC I and DMC III aerial imagery, provided as RGB in TIFF format. I have exterior orientation metadata as text, together with some interior orientation metadata (e.g. calibrated focal length, pixel size etc). I can import said data and metadata.

I have started by working with a block of >70 DMC III images and Align Photos to create a sparse cloud.

Unless I deliberately fix the focal length it ends up changing dramatically (>34000 mm vs. the calibrated 92mm). When I then use Model > Gradual Selection and progressively select points in the sparse cloud according to Reconstruction Uncertainty I get distinct structures forming:





When I fix the focal length the sparse point cloud appears thus:



These structures suggest systematic errors. I sense there is something wrong with the camera model? Is there something extra I should be doing to import Leica DMC images? Might the systematic errors be related to the fact DMC use multiple 2D arrays?

6

Camera Calibration / Clarification regarding Calibrate Fiducials

« on: March 20, 2020, 01:08:12 AM »

Am I correct in guessing that Calibrate Fiducials assumes the fiducial marks are symmetrical (equidistant from the principle point) and uses the given scan resolution to calculate their approximate coordinates? Is this done on a per-image basis, or use means (or some other statistic) if multiple images are selected together (therefore more representative coordinates)? I'm just seeking some clarification please. 

If this is correct it'd be great to add such to the manual for future reference.

Many thanks

Andy

7

Feature Requests / Calibrate Fiducials: um & dpi options

« on: March 20, 2020, 12:56:23 AM »

When using Calibrate Fiducials it would be wonderful to have the options to input the scan resolution in um (microns) and dpi please.

Many thanks

Andy

8

Feature Requests / 2D/horizontal (x,y) vs. 1D/vertical (z) markers and accuracy

« on: June 19, 2019, 05:38:04 PM »

There may be a way of doing this which I haven't yet discovered (please provide instructions if there is) but we really need a means creating 2D (i.e. horizontal) markers... i.e where only x,y coordinates were able to be measured. Also, where 3D markers are used (i.e. the only kind of marker I'm aware of in the current version) it would be very useful indeed to differentiate between 2D/horizontal (x,y) and 1D/vertical (z) marker accuracy, as the two are often very different indeed (e.g. when using GNSS to survey marker locations). Using just one overall Accuracy for each marker is very limiting when it comes to bundle adjustment with markers surveyed by different means or in different conditions.

9

Feature Requests / Ingest *.vrt and *.geom metadata for images automatically

« on: November 08, 2017, 01:28:40 PM »

The *.geom and *.vrt formats are becoming ever more popular for photo station ephemeris, camera models and RPC, not least in the airphoto sector. With 1000s of images having to be processed it is impractical to add this data manually. Can PhotoScan be updated to automatically recognise and import this data please?

Many thanks

Andy

10

General / Using *.geom for ephemeris and camera model for images?

« on: November 08, 2017, 01:25:08 PM »

I have a bunch of vertical and oblique airphotos I wish to use in SfM (*.jpg as it happens). Each one has a *.geom file, which appears to contain the photo station ephemeris data and the camera model for each. My question is, how can I automatically add all this metadata into PhotoScan. I have 1000s of images and doing it manually is not practical!

Many thanks in advance.

Andy

11

General / Using *.vrt for ephemeris/RPC of oblique airphotos?

« on: October 27, 2017, 07:37:17 PM »

I'm using oblique airphotos acquired by NOAA and made freely available as TIFF here:

https://geodesy.noaa.gov/storm_archive/coastal/viewer/index.html

Each TIFF comes with an associated *.vrt file... a format I'm not familiar with. To me it appears to contain ephemeris/RPC.

I want to generate geo-referenced point clouds, as well orthophotos from multiple TIFFs. Is there a way for importing the information from the VRT files into PhotoScan for this purpose (therefore with no GCP required)?

12

General / Copying markers between chunks

« on: April 29, 2017, 07:25:43 PM »

My project needs to be processed as several chunks. In the first chunk I have spent a lot of time carefully placing and naming a LOT of markers (non-georeferences, i.e. no XYZ coordinate, I just want to use them to align my chunks). How can I copy markers from one chunk and use them in another (granted, I'll have to place them all again in the photos, which is the whole point, but I won't have to carefully name each one over and over for each of my multiple chunks)...

13

General / Patches of high/low areas in point cloud of terrain

« on: December 20, 2016, 06:31:22 PM »

I'm using some vertical aerial photographs to create a Digital Elevation Model (DEM) and orthophoto of the terrain. I have ground control points (GCP) measured using Differential GNSS. The images align very well and the production of the dense point cloud (setting High) goes according to plan. However, in the point cloud there are gaps, where patches of ground are noticeably higher or lower than that surrounding it. It is hard to see without panning around, but when creating a mesh from these points the patches and their edges become really evident. This is nothing like the actual landscape. How are these caused and how can they be corrected?

Many thanks