Google Earth has been with us for ten years. It is the most complete virtual globe we have seen, and its satellite images are impressive. It is easy to get lost in remote regions that we probably didn't even know existed. However, the content in Google Earth is not exclusive to satellites. The 3D component in the images requires a bit more traditional solutions, and one of the main challenges its developers must face is not technical, but meteorological.
Google Earth is without a doubt one of the most important resources that Mountain View offers to the public. All you need is a stable Internet connection and a compatible browser (read Google Chrome), but that is only the first step. Its own engineers describe it this way: Google Maps helps you reach a destination, and Google Earth helps you get lost. The big question is how Google obtains and processes its content. Talking about "satellite images joined by algorithms" is simply unfair. If you want to know all the details, you should watch this video from Nat and Friends:
Granted, satellites do the heavy lifting, but the 3D results depend on images taken with airplanes. In other words, imagine a small plane equipped with five cameras, flying in a zigzag pattern. Visual recognition detects identical features in the photos, and when combined with GPS data, it is possible to calculate a depth map, leading to the 3D reconstruction and the application of textures. The images are edited in different ways, with light and color corrections, and even the removal of certain objects (cars, for example).
Still, the magic of editing cannot always counteract the power of nature. If the weather doesn't cooperate, flights must be delayed or split over several days, which happened when capturing images over London. When updating, the lowest frequency is recorded in large cities, with intervals of less than twelve months, and this translates into the excellent evolutionary map that Google shows us through Timelapse.
