A ground control point (GCP) is a point of known coordinates in the area of interest. Its coordinates have been measured with traditional surveying methods or have been obtained by other sources (LiDAR, older maps of the area, web map service).
Ground control points can be used in Pix4Dmapper to increase the absolute accuracy of the project.
Add / import coordinates of the GCPs
- The following step-by-step instructions are part of the whole procedure of how to include GCPs in the project. For more information about how to include GCPs in the project: How to include GCPs in the project.
- All GCPs must be in the same coordinate system.
The GCPs can be imported into the software using one of the following methods:
1. Using the GCP/MTP Manager: before running step 1. Initial Processing.
1. On the Menu bar, click Project > GCP/MTP Manager...
2. Select the GCP coordinate system: By default, the GCP coordinate system will be the same as the Output/GCP Coordinate System selected when creating the project: Step 2. Creating a Project. If the GCPs have a different coordinate system, select the correct coordinate system: How to select / change the Image / GCP / Output Coordinate System.
3. Import the GCPs coordinates. The GCP coordinates can be imported:
| Manually |
|---|
|
In order to import the GCPs one by one: 1. On the GCP/MTP Manager, click Add Point.
A new point is added on the GCP/MTP Table. This point is defined by default as Manual Tie Point (Type column) and is named by default mtp1 (Label column). When adding more Manual Tie Points, they will be named mtp2, mtp3, etc. 2. Edit the GCP properties:
1. Double click the Label cell of the point.
1. Double click the Type cell of the point.
1. Double click the coordinates cells of the point (X,Y,Z or Latitude, Longitude, Altitude).
1. Double click the Accuracy Horz/Vert cell of the point.
For more information about the GCP table properties: GCP / Manual Tie Point Table. 3. Repeat steps 1 and 2 for the rest of the GCPs. |
| From file |
|
In order to import the coordinates from the GCP file: 1. Make sure that the GCP file has the correct input file format. For more information about the file format: Input files.
The Import Ground Control Points pop-up appears.
3. (optional) Select the Coordinates Order: By default, X, Y, Z (Latitude, Longitude, Altitude) is selected. It is possible to change to Y ,X, Z (Longitude, Latitude, Altitude). For more information: Input files. 4. Click Browse... to navigate to the GCP file. Warning: The default value is 0.020 m, DO NOT modify the accuracy, unless the following article is read: GCP / Manual Tie Point Table.
1. Double click on the Accuracy Horz/Vert cell of the point.
For more information about the GCP table properties: GCP / Manual Tie Point Table. 9. On the GCP/MTP Manager window, click OK. |
2. Using the rayCloud: after step 1. Initial Processing has been completed.
1. On the Menu bar, click Project > GCP/MTP Manager...
2. Select the GCP Coordinate System: How to select / change the Image / GCP / Output Coordinate System.
3. Open the rayCloud: From the Menu bar, click View > rayCloud. The rayCloud opens and the Automatic Tie Points appear.
4. On the 3D View, click on one point that approaches the GCP position.
- On the right sidebar, the Images section displays all the images on which the clicked (selected) point is visible.
- On all the images the following appear:
- Green cross: Reprojection of the 3D point on the images.
- Orange cross: Position where the associated 2D keypoints have been automatically detected.
- Orange circle: The radius of the orange circle indicates the size of the area that has been taken into account to detect the keypoint.
5. On the right sidebar, in the Images section, the GCP area should appear in the displayed images. If not, click another point on the 3D View that approaches better the GCP area.
6. Click 
New Tie Point.
The Selection section in the right sidebar displays information about the New Tie Point:
7. Edit the point's properties:
- (optional) Change the Label of the GCP:
1. Click the Label cell of the point.
2. Type the desired name.
- Change the Type of the point:
1. From the drop down list select:
- 2D GCP: If only the X,Y or Latitude,Longitude coordinates of the GCP are known.
- 3D GCP: If the X,Y,Z or Latitude, Longitude, Altitude coordinates of the GCP are known.
- Check Point: If the X,Y,Z or Latitude, Longitude, Altitude coordinates of the GCP are known. A Check point is not taken into account during processing and indicates the accuracy around the areas where it is defined.
- Insert the coordinates of the GCP (for 2D GCPs and 3D GCPs):
1. Click the coordinates cells of the point (Latitude, Longitude, Altitude or X,Y,Z).
2. Type the corresponding coordinates.
- (optional) Insert the Accuracy (for 2D GCP Horizontal Accuracy, and for 3D GCP Horizontal Accuracy and/or Vertical Accuracy):
Warning: The default value is 0.020 m, DO NOT modify the accuracy unless you have read: GCP / Manual Tie Point Table.
1. Click the Horizontal Accuracy / Vertical Accuracy cell of the point.
2. Type the Accuracy of the point:
- For very accurate GCP, insert a low accuracy value.
- For non accurate GCP, insert a high accuracy value.
For more information about the GCP table properties: GCP / Manual Tie Point Table.
8. Click Apply.
9. Repeat steps 4 to 8 for other GCPs.
Mark / click GCPs on the images
The GCPs can be marked in the software using one of the following methods:
1. Using the rayCloud: when GCPs refer to a known coordinate system and after step 1. Initial Processing has been completed.
1. Open the rayCloud: On the Menu bar, click View > rayCloud.
2. On the left sidebar, in the section Layers, click Tie Points, then click GCPs / MTPs. The list of GCPs is displayed:
3. Select a GCP in the layer GCPs / MTPs: the right sidebar displays its properties and the list of images in which it is visible.
The estimated GCP position appears on the images with a blue circle with a dot in the middle. This position corresponds to the projection of the input 3D coordinates of that GCP in one particular image.
4. Navigate on the images to find the exact location of the GCP:
- Zoom in: Moving the mouse scroll wheel forwards.
- Zoom out: Moving the mouse scroll wheel backwards.
- Pan: Using the mouse left clicking.
- To change the size of the images: Move the thumb of Image Size bar.
5. Mark the exact position of the GCP on at least 2 images using the left mouse click.
- When marking a GCP on the images, all marks of this GCP are used to compute a new 3D point. At least 2 images need to be marked in order to compute the estimated 3D position of the GCP. This estimated 3D point is then reprojected in all the images where it might be visible in.
- The clicked position appears with a yellow cross and circle. The size of the yellow circle indicates the zoom level at which the marking has been done. Points that have been marked on a high zoom level are taken more into account than points that have been marked on a low zoom level.
- After marking a GCP on 2 images, a green cross appears on all images. The green cross represents the reprojection of the estimated 3D point and its location depends on the marks already made for the GCP.
- After marking a GCP on 2 images, a pink circle appears when the point is wrongly marked on other images. Outliers are indicated with pink circle and they do not influence the calibration results.
6. (Optional) Click Automatic Marking to automatically detect the position of the GCPs on the rest of the images.
7. When the green cross is at the correct position in most images, click Apply.
8. Repeat steps 3. to 7. for the rest of the GCPs.
9. When all GCPs are marked on the images, click Process >Reoptimize. This reoptimizes the reconstruction using the GCPs.
10. (optional) To Generate a new Quality Report, click Process >Generate Quality Report.
2. Using the Basic GCP/MTP Editor: when the GCPs coordinate system is Arbitrary or before running step 1. Initial Processing.
1. On the Menu bar, click Project > GCP / MTP Manager...
2. On the GCP/MTP Manager, click Basic Editor..., the Basic GCP/MTP Point Editor window appears.
3. On the GCP/MTP Table, select the GCP to be marked.
4. On the Images section, sort the image list:
By name: The images are sorted alphabetically.
By Distance to GCP: Images closer to the GCP are listed first. This sorting can be used only if the images and GCPs are geolocated using a known coordinate system. This is the default sorting mode.
By Distance to the Marked Image: When an image is selected and a GCP is marked on it, this option lists the images that are closer to the currently selected image. This sorting can be used only if the images are geolocated.
5. On the Images section, select the image on which the GCP should be marked. After the selection, the image is displayed in the Preview section.
6. Navigate in the image by:
- Zoom in: Moving the mouse scroll wheel forwards.
- Zoom out: Moving the mouse scroll wheel backwards.
- Pan: Using the mouse left click.
- Zoom in quickly on a specific point in the image: point with the mouse the point and press the Alt key. When the key is released, the view returns to the last zoom level.
- Zoom out to preview the zero zoom level: Press the Shift key. When the key is released, the view returns to the last zoom level.
7. Left click on the image to mark the point.
8. (optional) If it is needed to remove a mark, click 
Remove.
9. On the Images section, select the next image on which the GCP should be marked.
10. Repeat steps 5 to 9 for other GCPs.
11. Click OK.
Minimum how many GCP'S are required to.import in project
Hi Sharath,
A minimum number of 3 GCPs is required but we recommend including at least 5 to 10 GCPs. When the topography of the area is complex adding more GCPs could lead to a more accurate reconstruction.
Is it possible to process a project that contains GCPs on the ground around a building and also GCPs on the side/facade of the building? I wish to take a series of photos at oblique angles and several heights in order to later create an orthomosaic facade but I want to be sure that we can use targets on the side of the building with known coordinates in the processing part of the project. Thank you.
Hi Sebastian, it is possible to use the GCPs on the facades of the building. This way you will also add more information to the project and will be able to validate that the reconstruction is accurate.
As for the dataset, we highly recommend taking images that are directly facing the facade that you plan to reconstruct. We noticed that this way the results are better.
Best,
In the Basic GCP/MTP Editor, how many images need to be marked for each GCP?
Hi Alan, the minimum is to mark each GCP on 2 images. However, I would recommend marking it on more images (5 - 10).
Using the Basic GCP/MTP Editor is not as convenient as marking the GCPs in the rayCloud so I would recommend marking a few GCPs on a few images and reoptimize the project. After that, the position of the GCPs and the model should be aligned and you can continue to mark them in the rayCloud.
Best,
Is GCP importing allowed in the trial version? I am testing this software against Drone Deploy and it appears that the GCP functions are greyed out in the RayCloud editor. It may be because the Point Cloud and Mesh is still processing.... Please advise
Hello David,
The trial version allows all of the functionalities the same as the paid version. GCP importing and marking is therefore allowed.
Anyway, the GCPs are imported in the GCP/MTP manager, not in the rayCloud. It is always greyed out if the GCPs have not been imported.
The GCP/MTP manager is:
Regards.
In the Images Section on the right sidebar, is there a way to toggle to the next image using a keystoke instead of using the slider? This would make image marking much faster for me if there is.
Hello Blair,
For now, there is no way to toggle to the next image using a keystoke, the only way is by using the slider.
Regards.
Hi, this may be a very basic question, but what does marking the GCPs achieve? as in what does marking more GCPs achieve?
HI Vivek. Marking the GCP tells the software where the expected location of the point is within the image frame. This tells the software essentially what part of the model corresponds to the real world coordinate.
Hello, I am stuck here. I can't mark or click on the GCPs as nothing is appearing.
Hi there. It may be due to the fact that you are logged in as a discovery license. this has limited functionality. Do you have an active trial or license you can use?
Can I use Mavic 2 Pro and manual ground control points to get the topography of construction zones?
Hi Abdelali. Images from the Mavic 2 Pro are compatible with Pix4D mapper and ground control points are a good way to ensure the best accuracy.
Hi~
I am new in photogrammetry. This might be a basic question: When I import the GCPs, it alway show me
"One or more GCPs lie far outside the area covered by the images. Check if the correct GCP file and coordinate system are selected."
But I set the coordinate as same as my out put coordinate. What could be the problem?
Also, the Z value are quite different from the z value I documented by GPS in the field when I export terrain class as xyz file and load in QGIS to compare the data I collected by GPS.
For example, at the same location the Z value in GPS shows 29 but for the z value I created from point cloud is 8. I guess it happened because the elevation were recorded as ellipsoid, but I am not sure how to convert them to the same datum. (Also, would you please tell me where I can find the information of what type of height dose Pix4D export?)
Thanks~