The Quality Report includes error messages to help you easily identify what are the causes of inadequate output results, so that you can make the necessary corrections before rebuilding.
Below are listed the most common error messages and some recommendations on how to handle them:
Input: Photos and Collections
| Field | Message | Recommendation |
| Number of Photos | Fewer than 20 photos in collection. / Fewer than 6 photos in collection. | Check whether a collection was mistakenly split into two collections, and if so, merge the split groups into a single one. Homogeneous groups of photographs, all taken by the same physical camera with identical focal length and dimensions can be placed in the same collection. |
| Image Width | Missing property value. / Invalid property value. Value must be greater than 0. | This information can usually be obtained from your data provider or camera manufacturer, and then entered in the collection property sheet or Excel/XML photo list file. |
| Image Height | Missing property value. / Invalid property value. Value must be greater than 0. | This information can usually be obtained from your data provider or camera manufacturer, and then entered in the collection property sheet or Excel/XML photo list file. |
| Sensor Width | Missing property value. This property is necessary in calculating focal length. | This information can usually be obtained from your data provider or camera manufacturer, and then entered in the collection property sheet or Excel/XML photo list file. If you don’t have access to this property value, focal length can instead be set based on the pixel size and focal length values or the focal length 35 value. |
| Pixel Size | Missing property value. This property is necessary in calculating focal length. | This information can usually be obtained from your data provider or camera manufacturer, and then entered in the collection property sheet or Excel/XML photo list file. If you don’t have access to this property value, focal length can instead be set based on the sensor width and focal length values or the focal length 35 value. |
| Focal Length | Missing property value. / Invalid property value. Value must be greater than 0. | This information can usually be obtained from your data provider or camera manufacturer, and then entered in the collection property sheet or Excel/XML photo list file. If you don’t have access to this property value, focal length can instead be set based on the focal length 35 value. |
| Focal Length 35 | Missing property value. This property is necessary in calculating focal length. / Invalid property value. Value must be greater than 0. | This information can usually be obtained from your data provider or camera manufacturer, and then entered in the collection property sheet or Excel/XML photo list file. If you don’t have access to this value, focal length can instead be set based on focal length together with either pixel size or sensor width. If none of the required values are available, set focal length 35 to the default value of 50. |
| Principal Point (X,Y) | Principal point is too far from photo center. | Verify that the principal point reference’s correct origin and units were selected in the collection property sheet or Excel/XML photo list file. |
| Input Coordinate System | Missing collection’s coordinate system information. / Invalid coordinate system information. | If the XY values seem to be in degrees, i.e., they are in the range from -180 to 180 with 5-6 decimal places, the coordinate system is most likely WGS84 with EGM96 vertical datum. Enter this information in the collection property sheet or Excel/XML photo list file. |
| Vertical Datum | The WKT of one of your project coordinate systems references a vertical datum (GTX) file that was not found on your computer. | Download the required GTX file. See here for more information. |
| Name of Photo Collection - Photo Position Data | Missing XYZ position data. | Check if this information is available in an external positions file. If not, you can estimate the missing values by using the values of the adjacent photos, and enter this information in the photo property sheet. |
| All of the project’s photos have the same XYZ values. | Delete the position data from the property sheet. Verify if this information is available in an external positions file. | |
| Many of the project’s photos are positioned underground. | Verify that the correct geoid/ellipsoid are set in the photo’s coordinates system. Also check that the correct altitude mode (relative/absolute) and offset are set in the collection property sheet. | |
| There are more than 10,000 photos in the project. | Verify that PhotoMesh is running on a sufficiently powerful machine. PhotoMesh can process projects with more than 10,000 photos as long as the machine it is running on has sufficient compute, memory, and storage capabilities. | |
| Name of Photo Collection - Photo Orientation | Invalid value. Orientation values must be in degrees. | If values are in radians and photo information was imported in an Excel file, verify that the Omega/Phi/Kappa Units property is set to “Radians” in the Excel. Otherwise, If orientation values are in gon or radians, convert to degrees, and reimport the photos. |
| Name of Photo Collection - Band Definition | Mismatch between the number of bands according to the property sheet and the photo files’ metadata. | Fix value in property sheet. |
Input: Lidar
| Field | Message | Recommendation |
| Coordinate System | Missing Lidar’s coordinate system information. / Invalid coordinate system information. | Verify the validity of the WKT in the coordinate system dialog. Fix as required. |
| Location (bbox) | Lidar point cloud points are all outside the AT or reconstruction area | Verify in the Lidar List that the correct coordinate system and vertical datum are set. If a local coordinate system was set, it must be referenced to a known coordinate system. |
| Content | Lidar point cloud contains no points | Check your data. |
Aerotriangulation Result
| Field | Message | Recommendation |
| Triangulation Error | Triangulation error between 0.5px-1px. / Triangulation error > 1px | Check the sensor parameters. Verify that radial/tangential distortion values are correct in property sheet or Excel/XML photo list file. |
| Calculated Position | Difference between photos’ calculated and inputted position values exceeds the error tolerance defined by the aerotriangulation Calculation Mode parameter. All of the project’s photos have the same calculated position. | Verify that the control points were marked accurately and that photos’ inputted position and orientation information is correct. |
| Focal Calculated | Difference between photos’ calculated and inputted focal values exceeds the error tolerance defined by the Focal Length Accuracy parameter. / Difference between photos’ calculated and inputted focal values is greater than 50%. | Check the input value. Fix if required and then rerun the AT. |
| Principal point (X,Y) calculated | Difference between photos’ calculated and inputted principal point values exceeds the error tolerance defined by the Principal Point Accuracy value. / Difference between photos’ calculated and inputted focal values exceeds permitted deviation. | Check the input value. Fix if required and then rerun the AT. |
Control Points
| Field | Message | Recommendation |
| Number of GCPs | Only 1 ground control point was added to collection. | It is recommended to add at least 3-5 control points per AT tile, to enable them to be fully utilized in the AT process. When only two control points are provided, simple vertical adjustment, scaling and rotation are performed. When only one control point is provided, only vertical adjustment is performed. |
| Number of Photos Marked | Control points marked in 2 or fewer photos. / Control points marked in fewer than 2 photos. | Mark each control point in at least three photos. |
| Number of Photos Used/Dropped | Number of used is <= 2 | Add additional control point markings to the photos. |
| Geographic Error Horizontal (X,Y) | Per GCP: If greater than defined accuracy. / Per project: if >10% or GCPs have errors. | Check that the coordinate system and vertical datum of both the control points and photos are accurate. |
| Geographic Error Vertical | Per GCP: If greater than defined accuracy. / Per project: if >10% or GCPs have errors. | Check that the coordinate system and vertical datum of both the control points and photos are accurate. |
| Geographic Error Circular |
Per GCP: If greater than defined accuracy. / Per project: if >10% or GCPs have errors. |
Check that the coordinate system and vertical datum of both the control points and photos are accurate. |
| Triangulation Error RMS | Root mean square (RMS) exceeds 0.5 pixels. | Verify the accuracy of control point placement and their marking in photos. |
| Triangulation Error Maximum | Triangulation error exceeds 1 pixel. | Verify the accuracy of control point placement and their marking in photos. |
Maps
Photo Matches Map
Color-coded map that graphically depicts the degree of photo overlap. Each match (feature that was found in at least two photos) in the defined AT area is colored according to the number of photos in which it was found.
- Red – Minimal coverage (found in 2 photos)
- Yellow – Moderate coverage (found in 3 photos)
- Green – Strong coverage (found in 5 or more photos)
In oblique collections, moderate-strong coverage is generally necessary, but only in the area of interest, while on the project edges, minimal coverage will usually suffice. In nadir collections, minimal coverage even in the area of interest is generally sufficient. To improve matching in cases of weak coverage, it is recommended to use the AT: More Aggressive Match preset.
Connection and Projection Error Maps
Projection error - Median of the triangulation errors of all the features in each photo. The triangulation error is the distance in pixels for every automatically identified feature, between its original position and the position computed through aerotriangulation. Each camera symbol is color coded, based on the level of projection error. If the projection error is greater than 0.6 pixels in a group of images localized in the project, it could be an indication of a bad control point, constraints in the camera model geometry, or an area with challenging features (forested areas with large time and/or shadow differences, or extremely urban areas with inconsistent coverage). It is recommended to review that area in the project AT in order to verify that there are no errors on the input data (control point or images), and that the images have proper overlap and coverage.
Connection - Lines connect between photos with common automatic tie points, with the weight of the connecting line corresponding to the number of common points. 3D automatic tie points are obtained by computing the positions of features (I.e., characteristic points on photos) that were automatically detected in two or more overlapping photos. If the links between photos are weak, manual tie points can be used to establish a stronger correspondence between project photos and improve aerotriangulation results. If the number of links remains low, the AT: More Aggressive Match preset can be applied when building the project to increase the likelihood of matches being made. The color of the connection lines provides information about the strength of the connection:
- Black – Strong connections
- Brown – Weak connections
- Red – Initial connections that were later disconnected later in the AT bundle adjustment process. This indicates that the overall connection in the area was not strong enough to maintain the link between the photos.
Ground Resolution Map
Color-coded map that provides information on the highest resolution that was achieved in cm per pixel for each 3D automatic tie point within the defined AT area. 3D automatic tie points are obtained by computing the positions of features (I.e., characteristic points on photos) that were automatically detected in two or more overlapping photos. Make sure that the expected resolution level is achieved in the area of interest. On the project’s edges, lower resolution will usually suffice.
Distance Map
Graphically depicts the vertical and horizontal distance between every photo's original input position and the position computed through aerotriangulation. Each circle on the map represents a photo, with the size of the circle corresponding to the difference between the input values of the photo and its calculated values, i.e., the greater the difference, the larger the circle. The following are some of the distance map results you may see:
- A consistent shift in length and direction from the input to computed values (depicted by lines), generally in large drone projects - This often indicates a lag between the acquisition of the positioning information and the photo capture time
- A significant discrepancy between input and computed values in a specific stretch of the map, with minimal differences on either side. This may be caused by a consistent GPS offset between two different flights or by an incorrectly placed control point that has caused a significant shift in a specific section of the AT area.
- Arc between original and calculated values may be because of tangential effect. This most commonly occurs with rolling shutter cameras, and low-quality lenses.
To improve positioning in cases of significant discrepancies between inputted and computed positions, it is recommended to use the AT: More Aggressive Match preset.