Topics

1

General / GCP error (m) for each image

« on: August 16, 2017, 06:10:33 AM »

Hello,

I'm wondering if it's possible to get the error in meters for each image in a project from GCPs. I've identified at least 3 GCPs in each image, and can see the error of each GCP in meters and pixels (the GCP_errors figure attached). And the 'Show Info' for each GCP shows the error in pixels for each image (pixel_error figure attached). But I'd like to know a mean error for each image in the project in meters- is that possible?

Thanks,
Lauren

2

General / Camera/GCP error disagreement

« on: May 26, 2017, 05:04:35 AM »

Hello,

I'm working on a project and am getting weird results using GCPs to check the error of my model that has been georeferenced using 2 different camera datasets. I have 10 GCPs that I've selected in all images in another project, and exported them as an .xml file.

1. import images
2. align images
3a. georeference point cloud with camera position data A, error is 1 m, save as one project
3b. georeference point cloud with camera position data b, error is 3 m, save as different project
* At this point, everything seems to make sense, as camera position data A should be more accurate than position data B
4a. import markers (all unchecked), while the camera position error is 1 m, the markers check error is 2 m
4b. import markers (all unchecked), camera position error is 3 m, but marker check error is 1.5 m

I'm having trouble understanding why the camera positions with the lower error have a higher 'true error' (checked with GCPs), while the camera positions with the higher error have a lower 'true error'.

Any suggestions/ideas would be great, thanks!

3

General / Lens calibration not working

« on: March 13, 2017, 01:38:40 AM »

Hello,

I'm trying to calibrate several camera/lens combos (estimate f,c,K,P parameters) that I will be using to take images for photoscan projects. I'm using images taken of a large checkerboard, and following the workflow outlined here: http://downloads.agisoft.ru/lens/doc/en/lens.pdf

My problem is that I import ~30 images of the checkerboard, select Tools -> Camera Calibration, and then select OK (image attached of this window)... and nothing happens. The manual says that a dialog box should appear, and the calibration results will appear in the report window.

The only problem I can think of, is that in the calibration images, the checkerboard does not take up the full screen (example image attached), like the manual suggests.

Thanks for any help!

(I'm running photoscan version 1.2.6 build 2834)

4

General / Blue points (used matches) on image, but not aligned

« on: October 27, 2016, 03:22:26 AM »

Hello,

I'm trying to align one non-georeferenced image with ~100 georeferenced images of the same scene (a glacier). After aligning photos, the non-georeferenced image has points matched that are used in the point cloud (blue dots, shown in the attached figure), but the image (0807_010 in attached image) hasn't aligned with the georeferenced images (photo console also showed).

If points have been matched, I don't understand why the image does not align. I've also tried reseting alignment and re-aligning, which has not worked.

If anyone has an idea why the image is not aligning, or suggestions on how to align it (preferably not manually with markers, I'm hoping to keep the entire workflow automated), I'd really appreciate it.

Thanks!   

5

Python and Java API / Differentiating between Aligned and Non-Aligned Photos/Cameras

« on: October 21, 2016, 06:26:35 AM »

Hello,

I'm looking for a way in python to determine, after aligning photos, if certain photos aligned. Is there a command for determining this?

I've attached a screenshot of the Photos Pane, essentially I want to be able to write a loop so that if a camera in a list that I have does not align (camera 0313_002 in this example), I can skip the next step.

Thanks

6

General / Export Orthophotos in v 1.2

« on: October 19, 2016, 07:53:42 AM »

Hello,

I'm trying to use PhotoScan to orthorectify photographs (~100 of them). I can successfully orthorectify each of the images using the pre- v 1.2 technique (build orthomosaic, disable all but one image, export orthomosaic). However, when I try and utilise the Export Orthophoto tool in v. 1.2, only 2 of the photos will successfully export as orthophotos. This happens when I try and export all cameras as orthophotos, and when I individually select photos to export one at a time- only the same 2 photos will export.

I've made sure to save the project, and the 2 images that will export are similar sizes and qualities as the ones that will not export. They are all oblique images- but I'm confident that they're aligned correctly (most images are georeferenced, with an error of less than 1 m). I've also attached the console output from 1 of the images that will work with export orthophoto, and from 1 of the images that will not export.

I'm running 1.2.5 build 2680 (64 bit)

Any suggestions for using this new tool would be great, thanks!

7

General / Project combining archived/old and modern photographs

« on: October 07, 2016, 03:15:00 AM »

Hello,

I'm working on a project hoping to use old photos of a glacier to look at changes in it's size. I've also taken ~100 photos of the glacier this past year from a small plane (which are georeferenced with GCPs and with geotagged images from a GPS in the plane). 

I can successfully create a model of the glacier using photos from this year (2016). However, when I mask the ice from the 2016 photos and then include the old photos during alignment, no points from the old photos are included in the point cloud (despite used/blue points being shown on the photos (shown in attached images)).

I'm wondering if anyone knows why the aligned points shown in the old photos do not become part of the point cloud or the model. Or if anyone has suggestions for a workflow that may be more successful. I've tried the alignment with and without GCPs, with and without then image locations, and working with the 2016 and old photos in separate chunks and 1 chunk.

I've attached several photos- the generated point cloud, and 2 examples of old photos after alignment (with the used and unused points shown).

Thanks for any suggestions

8

Python and Java API / Using python to load multiple photos and apply individual masks

« on: August 09, 2016, 03:22:06 AM »

Hi everyone

I'm trying to load a subset of images in a folder into photoscan (DSC_1700.tiff - DSC_1750.tiff), and then apply a different mask to each image (DSC_1700_mask.png - DSC_1750_mask.png). The addPhotos line in the loop works, and the images are loaded. But the importMasks line is not working (I think every time importMask is called, it's applying that mask to all of the photos that are currently read in). Is there a way to specify the image to be masked? Or another way to loop through a subset of photos and apply an individual mask to each?

Thanks!

Code: [Select]

#!/usr/bin/python

# python script for photoscan
# reads in and masks specified photos (startphoto through endphoto)

import PhotoScan
doc = PhotoScan.app.document
chunk = PhotoScan.app.document.addChunk()

# loop to read in photos
startphoto=1700
endphoto=1750

for x in range(startphoto, endphoto + 1):
     chunk.addPhotos(['/Volumes/arc_01/FIELD_DATA/SFM/EOSS/PHOTOS_EOSS_20160330/106D800E/TIFF/DSC_%s.tiff' % x])
     chunk.importMasks(path='/Volumes/arc_01/FIELD_DATA/SFM/EOSS/PHOTOS_EOSS_20160330/106D800E/TIFF/DSC_%s_mask.png' % x,method='file',operation='union',tolerance=10)