Troubleshooting: Strip Align fails in dense forest areas

📌 Problem

Strip Align fails when processing a dataset collected over dense forest or heavily vegetated areas.

In the Strip Align log file, the final error may be similar to:

15:54:36.302 [ERROR] class std::runtime_error There are less observations 3260 than variables 6504.
More features are required for more consistent results.

🔍 Root Cause

In dense forest areas, the canopy is highly repetitive and does not provide enough unique features for the algorithm to identify sufficient tie points. When too few tie points are found, Strip Align cannot compute a solution and terminates with an error indicating that more features are required. 

This is similar to challenges encountered in photogrammetry workflows over dense vegetation, where image matching algorithms may struggle to find enough tie points between overlapping images.

🧠 What does the error mean?

The error indicates that Strip Align did not find enough tie points (observations) to perform the adjustment.

The specific observation and variable counts shown in the message will vary between projects, but the underlying cause is the same: there are not enough tie points available for Strip Align to run successfully.

✅ Probable Resolutions

1. Skip Strip Align

Strip Align  is not suitable for datasets consisting primarily of dense forest canopy.

If the project area is heavily forested and Strip Align reports that more features are required, skip the Strip Align step and continue processing the dataset.

No additional troubleshooting is required, as this is expected behavior for this type of dataset.

📚 Additional Notes

• This behavior is expected in dense forest environments.
• The error does not indicate a software defect.
• The error does not indicate a issue with the raw data acquired.
• The observation and variable counts reported in the error message will vary from project to project.

Keywords: Strip Align, dense forest, vegetation, canopy, tie points, observations, more features are required for more consistent results

📬 Need Help?

If you're still stuck, please Contact Support for assistance.