DroneDeploy provides tools to help you understand and improve the accuracy of your maps. Accuracy is generally divided into two categories: Relative Accuracy and Absolute Accuracy.
Q: What is the difference between relative and absolute accuracy?
A: Relative (or local) accuracy is the accuracy of a point on your map relative to other points within that same map. This is essential for measuring lengths, areas, and volumes. Absolute (or global) accuracy is the degree to which a point on your map corresponds to its actual position in a fixed coordinate system in the real world.
Q: What accuracy can I expect without ground control?
A: With a typical drone, you can generally expect a horizontal accuracy of approximately 1m and a vertical accuracy of 3m. For many projects, this level of relative accuracy is sufficient. However, if you need high absolute accuracy for property boundaries or construction design documents, you should use Ground Control Points (GCPs) or RTK drones.
Q: If I fly lower, will the accuracy be better?
A: Yes. Flying at a lower altitude increases the resolution of your images, which decreases the Ground Sampling Distance (GSD). Because DroneDeploy processes maps by matching common points between images, higher-resolution images provide more data points. This increases the probability of matching common points and improves the overall accuracy of the map.
Q: How can I further improve my map's accuracy?
A: To achieve the best results, consider the following:
Use a high-resolution camera.
Increase the overlap of your flight path.
Capture Ground Control Points.
Operate a drone with an RTK GPS system.
Q: Does the weather affect map accuracy?
A: Yes. High winds can cause motion blur or cause the drone to tilt, while poor lighting (overcast or high-noon shadows) can make it difficult for the software to find matching points between images.
Q: Can I use a lower overlap to save time?
A: Reducing overlap can lead to "holes" in your data or distorted imagery. For the best accuracy, we recommend a minimum of 75/75 overlap.
Further Reading
v2.2