Understanding Precision, Accuracy, and Error for High Accuracy Photogrammetry Survey

Accuracy, precision, and error are words often used interchangeably and can be confusing when trying to decipher what they mean. While some applications require high precision, others require high accuracy. When it comes to high accuracy photogrammetry (HAP) survey, it’s important to understand what each of these terms mean.

The dartboard principal is often used to describe the difference between precision and accuracy. When a player is aiming to hit the bullseye, this can be thought of as the reference or real value. Each dart throw could be thought of as taking a measurement. This measurement wants to fall on the real value or bullseye. After evaluating multiple throws, you can calculate the accuracy and precision of a player.

What Is Precision?

While precision is about the spread of the data, high precision is obtained by all throws (or measurements) being grouped together in a tight cluster. Even if they miss the dartboard completely, when the darts are grouped closely together, the precision of the throw is high. Variance quantifies the precision of a measurement which is found by calculating the difference between an individual measurement from the mean of all measurements.

What Is Accuracy?

Accuracy is about positioning the throws (or measurements) according to the bullseye (or reference). The spread of the measurements may be large but are still accurate when centered on the real value. When you average a series of measurements to find their mean, you can eliminate the variance and calculate the accuracy of the throws. This happens by finding the difference between the mean value of the reference and the throws.

What Is Error?

Error is discovered by the sum of the accuracy and the variance of a single measurement. Many surveyors use the root mean squared error (RMSE) for error distribution with approximately 68.3% of samples falling within +/- the value of the RMSE. Another popular measure of error distribution is the National Standard for Spatial Data Accuracy (NSSDA). This is similar to the RMSE except it encompasses 95.5% of the sampled points. By averaging multiple samples, the variance component of error can be eliminated to reduce the error of a measurement.

What’s More Important: Accurate or Precise Data?

So far, we know precision is about the spread of data and accuracy is the average position of the measurements from the reference point. But what is important for High AccuracyPhotogrammetry survey?

Two main types of measurement are common for HAP survey:

1.      Relative Measurements – the measurement of one point in a survey with respect to another point in a survey.

2.      Global Measurements – the measurement of a positioned measurement with respect to an external reference (i.e. British National Grid).

If you are interested in measurements with respect to an external reference, such as British National Grid, high accuracy is required. If you are merely interested in measuring a property of something relative to something else in a survey, high precision is needed. High accuracy and high precision aren’t always required and can affect the costs of a survey. Understanding the differences and which you need can help save you money.

Interested in learning more? At Insitu, we create innovative UAVs and technology solutions for HAP survey. Contact us for more information.