Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Messages - kfish

Pages: [1]
General / Re: DEMs constantly sloping up from one side?
« on: July 25, 2023, 10:51:05 PM »
Hi Alexey,

I am using the scale bar that is visible in the images to place markers and then create scale bars with known lengths in Agisoft from that. The DEM is then being constructed based on those scale bars. For example, see the attached picture below (one without the markers and another test on a different spot with the markers on it). Right now i only have scale bars in one section of the image, could this be causing issues? Is there a better workflow to follow for using scale bars to generate DEMs?


General / DEMs constantly sloping up from one side?
« on: July 25, 2023, 05:34:18 AM »
I've been trying to create DEMs of some small scale topography but I seem to be constantly getting issues where the DEM is showing a strong slope from one side of the model to the other. Attached DEMs of two examples of this. I've been following the process laid out in:

Bayley, Daniel TI, and Andrew OM Mogg. "A protocol for the largeā€scale analysis of reefs using Structure from Motion photogrammetry." Methods in Ecology and Evolution 11.11 (2020): 1410-1420.

It seems to work fine through all the steps but the resulting DEM has this weird slope issue which is preventing me from extracting information about the local features. Any ideas?

Hi all,

I've got a model that I created using SFM that I do not have GPS from the cameras nor do I have GPS of any ground control points. However, I do have an object of known dimensions (scale bar) from which I can provide x y z lengths in millimeters. Is there a way to input these known dimensions to then generate a DEM from? Or anything similar? Any help would be greatly appreciated!


Yes, I've attached the results from the Tools -> Survey Statistics option. Survey_1 and Survey_full show the Camera Overlap results. Survey_1_locations and Survey_full_locations show the results from the Camera Locations results.

If this is not the right thing to be showing, let me know. Still learning where all the options are within Agisoft.

Thanks for any help!

Hi All,

I am brand new to Agisoft as I just starting using it to construct orthomosaics and DEMs for my PhD research. I am working with a ~20 km x 8 km area that was surveyed at 15 cm resolution with 60% overlap. I managed to produce the orthomosaic without too much issue but then when I put together the DEM I realized that one region is reading far too high for elevation. When I went back to look at my camera alignment I noticed that many frames were showing elevation errors of up to 100 meters. The errors were so significant that is actual visible in the model of the cameras (see attached screenshot). I have other DEMs of the area and also know that it shouldn't be more than a few meters in elevation yet am getting far higher.

I don't have ground control points, so I am unsure on how to correct the error. I am have tried reprocessing the images and have both processed them as a complete set (all images from the 16 passes) and in small groupings of plane passes. The error is produced in both work flows.

I followed this tutorial to produce the products:,%20DEM%20%28without%20GCPs%29.pdf

Any ideas on how to fix the errors would be greatly appreciated! One thought I had was generating my own set of control points with lat/long/elevation based on the previous DEMs I have, but I am unsure if that would correct the error?


Pages: [1]