# 4. Spatial accuracy and Ortho-correction

VHR imagery has a high spatial resolution. But another measure of it's capability is it's spatial accuracy in terms of absolute geo-positioning and geometric correctness.

The most recent MAXAR satellites (WorldView-3 and 4) claim a pointing accuracy of CE90 <5m (circular error at 90th percentile is less than 5m). This means that 90% of points measured will have an error of less than 5m in the x and y direction (2D).&#x20;

This is quite impressive give the satellite platform is orbiting the earth at an altitude of 617 km.&#x20;

However, it is important to be aware that earlier satellites in the MAXAR constellation produced less accurate imagery. Quickbird-2 data claims a CE90 of <23m.&#x20;

The claims of geo-positional accuracy are very impressive, and we do often see this reflected in the data. However, it still pays to be wary and validate the accuracy against independent check data where possible.&#x20;

Furthermore, these claims are calculated by taking into account the terrain, however in practise, we may be working with data that has not been adjusted in this way.&#x20;

For more detailed explanations and information on how this is calculated, [read this white paper](https://dg-cms-uploads-production.s3.amazonaws.com/uploads/document/file/38/DG_ACCURACY_WP_V3.pdf).

### Ortho-correction

To ensure accuracy and geometric correctness, VHR imagery needs to be ortho-corrected.

Ortho-correction is the process of adjusting and resampling the raster image to the underlying terrain. This is normally done using a [Digital Elevation Model (DEM)](https://en.wikipedia.org/wiki/Digital_elevation_model), in GIS, photogrammetry or image processing software.&#x20;

{% hint style="info" %}
Ortho-correction may be done by the satellite provider, or by the end-user depending on the product sold and the level of processing applied. The important thing to know is whether it has been ortho-corrected, and if so, the source and quality of the input elevation source used in the processing. &#x20;
{% endhint %}

The image below shows the change in image geometry following ortho-correction of imagery. In this example image of the Barff Peninsula in South Georgia, there is a shift in the position of the high peaks of approx 170 metres.&#x20;

![WorldView-2 © 2003  Maxar Technologies.](/files/tGnzteoeTGAzUku4snOH)

{% hint style="info" %}
We can improve accuracy of ortho-correction by applying ground control points to the process. If ground control points are available, which are measured at a higher accuracy than the standard dataset accuracy, then we can use these to improve the georeferencing of the imagery
{% endhint %}

#### Ortho-correction quality

In areas of very steep terrain, and where the image acquisition geometry is far from nadir, it is not uncommon to see ortho-correction artefacts in geometrically corrected imagery. This is simply due to the viewing geometry of the satellite and the fact that it could not 'see' all parts of the terrain. Therefore, when the image is corrected and resampled to the terrain, it is 'stretched' and therefore blurred/smudged.&#x20;


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://guides.geospatial.bas.ac.uk/10-things-to-know-about-vhr-satellite-data/4.-spatial-accuracy-and-ortho-correction.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.