Accuracy V Quality.

It’s often a game of give and take and in photogrammetry, accuracy and quality are two key aspects that every operator wants to deliver to clients.

Their relationship is often a trade-off and depends on the reason for the flight in the first place. Sometimes, improving one can come at the expense of the other.

For example, a project requiring precise measurements (like civil engineering) would prioritize accuracy over visual quality. A project created for visualizations (like virtual tourism) would prioritise higher quality over accuracy make it more visually appealing.

Accuracy:

This refers to how close the model is to real world measurements of the area the subject.

There are many factors that affect accuracy such as camera parameters and image quality, using control points, using RTK and the processing software.

One of the ways that accuracy is measured is by giving it a RMSE (Root Mean Square Error) score. This is the final calculation of the accuracy of the aerial mapping outputs when all possible sources of error have been factored into the model (not just GSD or GPS error).

Quality:

This refers to the level of detail in the model.

Factors that can affect quality such as the number, resolution and overlap of the images ambient light conditions. Applications that need high quality models are filming and gaming, cultural heritage, and virtual tourism.

The best way to measure the quality of a model is by a GSD (Ground sampling distance) score. This is a measure of the distance between centre points of each sample taken of the ground. Since we’re talking about digital photos in drone surveying, each “sample” is a pixel. In simpler terms, the GSD represents the size of one pixel when measured on the ground.