Messages

1

General / Re: Ruler tool on 2D photo

« on: April 22, 2021, 12:07:55 AM »

Thank you very much for replying!


Just so I understand what you are saying -

When viewing the image and using the ruler function the program projects this point onto the sparse point cloud? (I have not yet made the dense point cloud) I am pretty amazed at the apparent accuracy of this as I have the ability to measure small objects - when I am pretty sure there would not be many tie points around it.

Apologies, I am not familiar with the pickPoint function API

2

General / Ruler tool on 2D photo

« on: April 21, 2021, 06:12:14 PM »

Hello!

We have aligned 360 degree panoramas together (many panoramas taken very close to each other) and the results are really good. We have given the model real world coordinates from known locations around the scene (accurate XYZ points known from pre-existing model).

Recently found that you can use the 'ruler' function to check measurements whilst viewing the panoramic photograph, rather than the point cloud - this is sometimes easier to see what you are clicking on.


My question is - how does Metashape know the XYZ location of the points you are clicking? (I see the XYZ is shown when clicking the first ruler point) Is it just using the tie points behind the image - if so, would the ruler function work better (more accurate) once a dense point cloud has been created, rather than the initial alignment?


I hope that made sense!

Thanks.

3

General / Re: Straightening spherical using Agisoft output

« on: September 10, 2020, 05:10:14 PM »

The project has been assigned known coordinates of the building from a 3D model which has its own coordinate system (eg corner of building has coordinates (0,0,0).

I am viewing the project side-on with the floor beneath the sphericals.

EDIT

I should say that I am aware that the sphericals are squint as they were not photographed straight. This means that the alignment was completely correct.
- I am trying to understand by what degree the sphericals are squint in a roll, pitch, yaw format so that they can be correct for use outside of Agisoft. It would appear from the model view that this information is stored somewhere.

4

General / Straightening spherical using Agisoft output

« on: September 09, 2020, 01:19:20 PM »

Hi there,

Was wondering if it was possible to use the post alignment rotation calculations to straighten a spherical image?

Dataset consists of multiple spherical cameras aligned together and given coordinates through known ground points (XYZ).

I see that yaw, pitch and roll are calculated in Agisoft, and can see that Agisoft displays that the sphericals are not quite straight due to the horizontal line in each spherical being squint. I was wondering if I could use any of the information generated by Agisoft to straighten these panoramas after alignment, rather than doing each image manually.

I have tried using the straight Yaw est (°), Pitch est (°), and Roll est (°) in hugin but they are not correct.

Image attached of a spherical camera being displayed as not quite level.


Any help appreciated!

5

General / Re: Read Only Mode Won't Let Me Save Changes

« on: September 19, 2018, 10:59:42 AM »

Save it with a different name

6

Feature Requests / Re: Align chunks by Photo

« on: August 20, 2018, 02:33:40 PM »

Surely you would need to have more than one photo in each chunk to do this with any kind of accuracy? I do a very similar thing but have at least 3 cameras the same in each chunk and it works pretty well.

7

General / Re: Problem with processing equirectangular photos from a spherical camera

« on: April 27, 2018, 04:35:45 PM »

By chance, I was playing around with this dataset and had almost the exact same workflow as kalevj.

Seemed to work reasonably well.

Not done an alignment of 360 degree equirectangular images along a transect before, only in a grid format - I believe that it does turn out a lot better in a grid format due to the increased chance of overlap.

Anyway, screenshot attached of my results.

Thanks for the dataset!

8

Python and Java API / Re: Masks_From_Color.py

« on: April 27, 2018, 01:55:53 PM »

Big +1 on this from me.
Being in Scotland there are many, many varying colours of grey, white, black and blue in the sky at one time... being able to mask all these through scripting would be incredible.

Have also been looking for a way to only apply these masks in the top 50% of the image (along the horizon) so as not to mask all the important stuff with the same colour in the bottom half...

This could be achieved by either applying the masking to only the top 50%, or by doing the 'subtract selection' at the end to remove the masks that have been made in the bottom half.

9

General / Re: Problem with processing equirectangular photos from a spherical camera

« on: April 27, 2018, 12:43:24 PM »

Yes, I also have just tried the script, and once you find the 'sweet spot' for the sky it works superbly well. Currently trying to adapt the script to give me back the option of only masking out that colour in the top 50% of the image. Would you be interested in having this option also? I can post it here if wanted.

10

General / Re: Problem with processing equirectangular photos from a spherical camera

« on: April 27, 2018, 11:57:07 AM »

Thanks! Masking off the sky worked.

Masking the sky automatically would be great. Would it be possible to automatically mask areas of the images by color ranges? Because the sky is usually quite contrasting compared to trees and buildings.

Right now, I used the "Magic wand" tool. On small datasets, it's great. You have to click on all the sky areas but don't actually have to draw the mask yourself. But on larger datasets, this method would become quite time-consuming.

Kalev

It is definitely possible to automate masking the sky in other software (we personally use Photoshop to do this for 360 equirectangular images). We basically set up an action and do this:
1. Cut the top 50% of the image into a new layer (along the horizon) - this means anything that happens to have the same colour as the sky in the bottom of the image is not masked out)
2. Open up the layer styles dialogue and then adjust the blending option sliders on grey (since I am in Scotland!) or blue to remove the sky.
3. set the background of the image to be black and the front image to be white (black parts work as the mask).
4. Merge layers back together.


Unfortunately this relies on a lot of factors (images taken with a straight orientation i.e. no squint horizons, and a consistent sky in all of your images). there are probably better ways to do this, I am currently investigating myself! You can also use colour selection tool in Photoshop - however again this relies on you having a very consistently coloured sky.

An addition in Photoscan to be able to do this would be hugely beneficial.. Even just a colour selection mask would be great. It would make Photoscan an all in one application for this process which would be incredible.

Will definitely have to try out the script.

Have a look at https://github.com/agisoft-llc/photoscan-scripts/blob/master/src/masking_by_color_dialog.py

P tried tjis script on your image set and it masks al the image!

11

General / Re: Using 360° panoramic Pictures in PhotoScan professional

« on: April 27, 2018, 11:41:43 AM »

Hello Phish,

Have been using 360 equirectangular images for a while in photoscan and there are a few things you should take into account.

The photos should be taken using the highest possible quality in the camera - this will obviously give you a lot more tie points and help a lot with alignment.

As suggested by Alexey...

However, I also suggest to apply the mask (could be the same for all the images) to cover the unused area in the bottom on the images.

I would recommend masking the bottom part of the images as it would decrease total alignment time and could increase accuracy. I would also recommend masking out the sky if taken outdoors. This is a very simple process if you want to use the same mask in all images - see below.

Hi scr1tch,

Here is the workflow.
1. Generate the mask you want for one single picture in PhotoScan using the SELECTION TOOLS in the main icon bar

2. Then TOOLS > EXPORT > EXPORT MASKS. Save as a single channel mask image.

3. TOOLS > IMPORT > IMPORT MASK. Choose these settings: FROM FILE, REPLACEMENT and APPLY TO ALL CAMERAS. You then have to enter the filename of the mask you stored in step 1, e.g. mask.jpg. Then press OK.

4. A window will pop up asking you to select the FOLDER where the mask.jpg file is stored in. Navigate to that folder an then click the SELECT FOLDER button.

DONE. Your mask should now have been applied to all images. Note that this workflow will apply the same mask to every picture, e.g. very useful to ignore 'overlays' in images.

Regards,
SAV

I also strongly recommend that you take the panoramas close together to increase the overlap. Depending on the size of the area, one 360 panorama every 2 meters should be enough.


CM