How to merge chunks with drone imagery into DEM and orthomosaic

[Jazzy]

Hi all,

I am new to the forum. I tried looking for an answer to my question before but could not find it, so I hope I am not asking previously answered questions.

I am trying to merge two chunks of about 275 multispectral drone images (so 2500 in total) into one taken from a DJI Mavic 3 Multispectral, but I am unsure how to do this properly. Particularly, which steps do you carry out within each chunk separately, and which steps do you carry out within the merged chunks. Also, the one unaligned image seen in the attachment is from a calibration panel.

I have already done:

1) Workflow -> Align Photos (All Chunks)
2) Workflow -> Build model (All Chunks)
3) Workflow -> Align Chunks (Camera-based)
4) Workflow -> Merge Chunks with Merge Assets after creating 3D models

I now want to create a DEM directly from these merged 3D models. However, even though I now have two 3D models in 'Merged Chunk',  with one in normal font and one in Italics, when I re-run Workflow -> Build DEM after merging, it still only creates a DEM from Chunk 1, and not Chunk 2.

Is it perhaps a better idea to run the whole workflow from A-Z for each chunk separately and use a different program for merging Chunks? I hope someone can help. Thanks a lot!

[Alexey Pasumansky]

Hello Jazzy,

You can merge two 3D models in the "merged chunk" by selecting both in the Workspace pane and choosing "Merge" option from the context menu after right-clicking on selection.

After that you can set the resulting model as active - again right click on the 3d model's label and choose Set As Default option from the context menu. The default model will be displayed in the Model view mode and all the model-related operations will be applied to it.

[Jazzy]

Hi Alexey,

Thanks a lot for the prompt and insightful answer. I think I managed to make this work!

However, if I may ask another question... I have to divide my drone flights into 2 chunks due to RAM limitations. However, when merging the DEM's resulting from processing both chunks individually, I notice a difference or 'step' in the altitude between the boundary of the parts merged from chunk 1 and chunk 2. The result is that elevation in part 1 (chunk 1) of the DEM is always a few meters higher in altitude than the other (chunk 2), which is incorrect.

I know that fixing this probably relates to a combination of aligning both chunks into a similar reference space, and/or camera optimisation, but I cannot find out from the forum or User Manual exactly how to do this most efficiently for large flights of hundreds to thousands of images. Do you have a preferred workflow that fixes this issue at least relatively efficiently?

Thanks a lot again for all the help so far!

Cheers,

Jazzy