OSINT Trick: Overlaying drone footage onto Google Earth

It is possible to overlay drone footage onto Google Earth as a way to get more contemporary images of an area.

One of the hardest quests in every investigation is obtaining high-resolution satellite imagery. Most people don’t have the budget for the top providers such as Maxar, and those that do, are normally limited in how much they can obtain.

This guide will show you how to bypass that financial obstacle with a nice feature in Google Earth called Image Overlays. This trick requires Google Earth Pro and a good image to work with.

In this case let’s use a screenshot of drone footage from Brovary, an eastern part of Kyiv. This image was published in The Guardian.

As you can see this image was taken from drone height rather than satellite height, so it is much more limited in the area of coverage. Also note that it is not a top-down view, which we will explain later.

The first thing we need to do is find where this image depicts. This is done by geolocation and hopefully, you can skip this part if you are using drone footage of a known area.

Below, we have found the location and oriented the compass to roughly match.

IT is important to start off with a similar view and make sure you are looking at the correct place.

Next, we want to add the image. Right-click on your desired project folder and select Add > Image Overlay. This will open up a window for you to select the image.

Right-click in the project folder, > Add > Image Overlay

Google Earth should look like it does below, with a new window appearing and green lines.

This is the main image overlay setup view, in the window that appeared, you have options to name the image, load it from a file or link, add descriptions and control the settings.

This new window allows you to name the layer and also import the image. In the Link section, hit Browse and select the image you want. It should appear over the map like below.

The image will load like shown above, in the green markers. These markers can be used to position and scale the image. By dragging the cross in the middle, you can move the image. The corners allow you to scale and the diamond allows you to rotate.

This is obviously no good so we can use the green markers to adjust the image to the correct scale. you can also rotate it with the green diamond shape. Adjusting the transparency may help with this.

Try and match the corners first and remember if the drone wasn’t taking an image straight down, there is a good chance the image will look really warped.

The image with features lined up. Remember to line up the bottom of buildings rather than the tops, because the viewing angle is so low. Other features to match up are corners of walls and road markings, as they do not change based on viewing angle.

I like to also add details in the properties such as links and information because it may be useful in future. Importantly it allows you to trace the sources of your information. If you are conducting a serious investigation, now is the time to archive the source if you haven’t already.

Add in any details such as image source in case you need this for future.

It’s not perfect, but here is the result:

By adjusting the camera view to a rough approximation of the drone position, the image is a lot clearer. This can be good for presenting findings on a map.

This can become very useful for investigations if you need to present the findings in a visual way.

Of course, the quality of the input makes a difference. Due to drone camera angles, it normally doesn’t suit a top-down view when looking in Google Earth.

The next hurdle is dealing with the field of view. The Google Earth Image overlay feature only appears to allow rectangles. You cannot warp the image into other shapes which could account for certain fields of view.

Seen here the main road and tall residential blocks match this drone footage of Mariupol, but the top right corner shows the waterline is way off. We would need to Warp the image for it to match properly.

One fiddly solution is to go into a complete top-down view in Google Earth, export a HD image then load it into an image editor.

Try to export a large image file so you have fine details to align with.

Once in an image editor, you can load the drone footage screenshot and warp it. Here I am using Affinity Photo 2, but options like Photoshop or GIMP (free to use) are also good.

Loading the image into Affinity Photo 2.

This will make a very ugly shape that looks awful from above, but the goal is to have all corners matching so that on Google Earth when you look from the drone’s height and angle, you can see its view.

Using the warp, you can adjust for viewing angle. This will typically reduce the clarity of distant objects.

Now you can export the image without the Google Earth base layer, so you have a transparent background:

Remember to remove the background map and ensure the background layer is transparent.

This is slightly better but still not amazing:

The result is better but not perfect. With ample time it could be further optimised, but this suffices for my needs.

As you can see, there are definitely limitations. Some of these can be solved in other GIS software applications such as the free QGIS, but that is outside of the scope of this article.

Of course, our trick can be done with more than just drone footage. In fact, if someone has screenshots of satellite imagery you can use that too. One example is the Maxar imagery that has been purchased by news outlets or even released by the company itself. While this is not georeferenced, it can be overlayed onto Google Earth in the same manner and function as if it was!

This example shows a screenshot of Maxar imagery, it is much easier to align. The same would be true for high-altitude drone footage that looks more directly down at the target location.

Again the slight difference in view angle (due to satellite position) can make the image slightly off in some places, but it is much more aligned than low-angle drone shots.

As far as analysis goes, this isn’t a dealbreaker for me as I am likely toggling the imagery anyway.

If you liked this or have other cool solutions to this, hit me up on Twitter @tomtomjarvis.