Hi everyone,
I’ve created a new topic because in the old one maybe I wasn’t very clear. My problem is that when I add the GCPs (created in WEBODM) the orthophoto and DTM files are empty (but the 3d model is scaled, so the GPC file is correctly read), if I don’t add the GCPs they are generated. I’ve tried to change every parameter, but I couldn’t obtain a good result. I tried with datasets created with different cameras, so I’m pretty sure that the problem is not in the exif files (always with ground images). I know that I could export the 3d model and generate the dtm and something similar to an orthophoto in CloudCompare, but I’d really like to resolve this problem to obtain the maximum from WEBODM.
Thanks
Maybe you did this on the other thread, but problems such as this are usually easier to solve with a shared imageset and GCP file.
Hi and thank you for answering. I attach a link to a folder with part of the photos of one project and with the GCP file that I used. Thanks.
dataset
I insert a link to my console, for the dataset that I’ve shared, and another one to my 3d model. Maybe someone can explain me what I’m doing wrong. Thanks!
3d model
CONSOLE
I willl get to you. Folks are following up on these things in their spare time . In busier weeks, that followup is slower.
No problem, I get it! I’m just trying again and again to resolve this problem by myself, so I keep the thread upgraded! Anyway, I’m pretty sure that I’m doing something wrong and that is not a problem of the software. Thanks for your help!
It seems that you try to process a small object where GCPs are not very useful. I processed your dataset using the 3d-object-preset without a problem. Here is the result shown as a textured 3d-model.
Hi Hubertus and thank you. I’ve tried also with a larger dataset, but the result was the same. With all datasets that I’ve used, I’ve obtained textured and scaled 3d models, the only problem is in the raster files, the dtm and orthophoto, and I don’t know why, because the GCP file is correctly read. So, do you think that if the area is too small is not possible to generate orthophoto and DTM?
Thanks
You are right, the area is too small. You have only 5 gcps, that is probably not enough to calculate the transformation. Why do you need the georeferenced images?
Mmm, but the 3d model is georeferenced (I linked it in this thread), in the 3d view mode I can click on the little map on the left and if I do this, I see the model set in Ventimiglia (Liguria, Italy). I can also choose the top view and see the model from the top and also the measurements are right, so the GCP file is correctly read. The only things that I can’t generate when I add the GCP file are the orthophoto and the DTM. I could export the model and elaborate the DTM in CloudCompare, where I could also export a top view…but I’d like too understand why I have this problem in webodm. I use the scaled images to produce technical designs (I’m not sure if this is the right expression in English). It’s not a problem, because this is an old work (I did it using photoscan), but I’d like to use more and more webodm from now, because I think it has a lot of potential, so I want to understand what I’m doing wrong. Thanks!
Hi everyone! Great improvement! I’ve changed some parameters (I’ve inserted the txt of the console in Drive, it’s called "console_orthophoto) and I’ve obtained a perfect georeferenced orthophoto! Now I have to understand what parameter has made the difference and then I’ll report here. I haven’t resolved yet the problem with the DTM, but I’m confident.
I got it! The problem was the crop parameter, I’ve disabled it and now everything works perfectly! Thanks for your time guys, hope that this thread can help someone else in the future!
Ha! Fantastic!
Excellent. Thanks for following up.