This article provides step-by-step instructions on how to merge projects in PIX4Dmapper. There are two parts when merging a project: Before merging projects and while merging projects.
It is possible to merge projects for which step 1. Initial Processing has already been completed in PIX4Dmapper. This allows the combination of information from different projects into one single project.
Merging projects is useful when:
- There are different image acquisition types, such as terrestrial, circular, and grid image acquisition types are used. This is very helpful when reconstructing an object which requires a terrestrial, circular, and grid images. For example, images obtained from a handheld camera and a drone can also be combined to generate detailed 3D models.
- The dataset is considerably large for the capabilities of the available processing resources.
Merging projects can speed-up step 1. Initial Processing time. It can also be used for projects having processing issues during step 1. Initial Processing with respect to memory use (not enough memory).
Before merging projects
- Create a new project for individual flights following: Create a project. We will call this the subproject.
- Process Step 1. Initial Processing for each subproject.
- Identify and mark a minimum of three manual tie points (MTPs) that have the same name and point to the same feature on both or all the subprojects. How to add / import and mark manual tie points (MTPs) in the rayCloud. It is very important for the MTPs to have the same name at the same location in all subprojects. If the naming is not consistent throughout the subproject, it can cause processing issues. Ensure the MTPs are located where the subprojects overlap.
- Reoptimize each subproject after adding MTPs and save them. Menu Process > Reoptimize
If a project is too large to be processed at once due to the processing resources, it can be split into subprojects.
- If the project has image geolocation, the project can be split using the Splitting Projects into subprojects options: Processing Large Datasets.
- If the project does not have image geolocation, the project can be split manually: How to split a project into subprojects
- The coordinate system (both horizontal and vertical) of the images has to be the same in all subprojects. Note: A subproject with no image geolocation can be merged with a subproject with image geolocation.
- The GCPs must be in the same coordinate system (both horizontal and vertical) in all subprojects. Note: A subproject with no GCPs can be merged with another subproject with GCPs.
It is highly recommended the GCPs to be added in the merged project, especially when more than 2 subprojects are merged. - The output coordinate system (both horizontal and vertical) of all subprojects has to be the same. Note: A subproject with no georeference (arbitrary system) can be merged with a subproject with georeference.
For more information about the Image / GCP / Output Coordinate System: How to select / change the Image / GCP / Output Coordinate System.
Merging projects
1. Open PIX4Dmapper.
2. On the Menu bar, click Project > New Project..., the New Project wizard opens:
3. In Name, type a name for the merged project.
4. (Optional) In Create in: click Browse... On the Select project location pop-up, navigate to select the folder where the project and results will be stored and click Select Folder.
5. (optional) Select the check box Use As Default Project Location to save all new projects in the selected folder.
6. Selected the option Project Merged from Existing Projects.
7. Click Next.
8. On the Merge Projects page, the Duplicate camera parameters option is activated by default. If the camera model of the different projects is the same, it is duplicated by adding the name of the project at the end of the original camera model name.
9. Click Add projects...
10. On the Select Projects pop-up that opens, browse the project files. Select the projects to be merged (multiple selection is possible) and click Open.
11. Repeat step 9 and 10 to add more projects (if the processing resources can handle multiple projects)
12. Click Next.
13. Click Finish to create the merged project.
14. Automatically (no user intervention), the subprojects are combined and processed together. The final result of step 1. Initial Processing is generated.
15. Verify the Quality Report. One block should be created.
16. When the reconstruction consists of a single block, use the results to create another merged project, or start step 2. Point Cloud and Mesh and 3. DSM, Orthomosaic and Index.
- If at least 3 Manual Tie Points were marked in the overlapped area between the subprojects in each subproject, and if they have the same label in all subprojects, a pop-up appears when creating the merged project, asking if they are the same and have to be combined. If they are the same, then the software can join the blocks and create only one block.
- If there are not at least 3 Manual Tie Points when merging, two or more blocks are created. Therefore, after merging the projects, it is required to add at least 3 Manual Tie Points in the overlapped area and Reoptimize in order to obtain a single block. By generating the Quality Report it is possible to verify that one block was created.
Troubleshooting
It is very important that before merging, all subprojects are processed successfully and there are no issues reported in the initial checks of the quality report, e.g. bad camera optimization, uncalibrated camera, or multiple blocks.
If possible, try processing projects with similar flight height and camera orientation together as one project rather than merging. This can simplify the process of getting the sparse point cloud to align properly.Cases where the projects fail to align properly relative to each other are due to:
- There is not enough overlap between image acquisition plans.
- Make sure that each plan captures the images with enough overlap.
- Not enough matches between the subprojects.
- Add more common Ground Control Points or Manual Tie Points in the subprojects before merging.
The workflow for merging multispectral projects is different and is described here.
Excelente explicación, de mucha utilidad en los levantamientos para catastro urbano en donde las cantidades de fotografías son enormes.
I am trying to model a bridge. I took a series of drone shots which are georeferenced and I built a model from that. I also took non-georeferenced ground shots from a Sony camera and built a model from that. The scale on the ground shots is much smaller.
Do I need to scale that project before merging and/or does it have to be georeferenced as well or will the MTPs take care of referencing and scaling?
Hi Tylor,
You don't have to scale your project, MTPs will take care of the scale. Your final scale will be the same as the project with drone shots that are georeferenced.
Regards,
Hello,
I am having problems merging two projects and I am not sure where the issue lies. I have merged projects before successfully so I am not sure what I am doing wrong. My first project is nadir and my project is oblique using an IPhone. When I merge the projects, they end up on two different planes within the merged project. Do you know what could be causing this issue? The projects on their own processed fine.
Thanks
Hi!
The best solution for you would be to post on our Community where you can attach the Quality Reports so we can investigate the issue. For our convenience, we would need three; nadir, oblique and merged report. We look forward to your post and the data so we can start the troubleshooting!
Regards
Thanks for the guide.
When I merge projects, this is my workflow:
1. Run step one for each subproject.
2. Mark GCPs in each subproject, and then reoptimise.
3. Create a new project by merging the subprojects.
Once the project has finished building, the GCPs show as marked in the raycloud but not in the quality report. To get a geolocation accuracy number, I have to reoptimise the merged project. This takes considerable time, as the project is now large.
Can I skip this reoptimisation, relying on the quality reports from the subprojects or is this reoptimisation recomended?
Hello, you should not need to Reoptimize, the GCPs should be taken into account. Please create a new community topic here
describing the problem and attaching the quality reports of the individual projects (before merging) and the quality report of the merged project (before the Reoptimization), so that we can better understand the case. Thank you.
I also had to rematch and optimize in order to get this to work.