How does Google do that? – Part 1
Remember family trips, back in childhood!
Adults gather to plan, outlining possible destinations and the paths to take on a map. The sound of numerous arguments on the best route to take! The sound of crumpling map paper – which eventually, tears apart in a fight amongst the children. Now, where? Lost?
Well, technology had plans to cure such problems.
At this point, you are probably like what is the difference?
Well, Google Maps provides us with excellent turn-by-turn navigation, keeping our data synched across devices, which might be the most convenient option for an on-the-go service. Whereas, Google Earth offers additional 3-D content. It also allows to use special controllers such as Space navigator and the LEAP motion. These aid to dig deeper for the more intricate details and experience simulation.
Back, when I was new to Melbourne, I set out to see the city – alone. What was meant to be a casual stroll, turned out to be a frightening venture. Within seconds from getting of a train, I was lost. Didn’t know where to go or whom to ask! I truly credit Google Maps for saving my lost soul that day. (From then on, made sure I had company)
We all secretly share a lot of gratitude to Google for providing us with a saviour!
So, how exactly does Google do this?
It is a mixture of many complex processes and algorithms, used to analyse each step. In short it’s a mix of sources like StreetView, aerial imagery and information provided by users.
Google Street View:
Street View plays a major role in capturing images using googles fleet of equipment to get the top imagery. This includes navigating from the narrowest of alley ways to the widest of freeways, from the hottest of deserts to the snowy caps on mountain peaks. This wouldn’t be possible without the following equipment:
-Street View car
-Street View Trolley , a system developed to fit through museum doorways and navigate around sculptures.
-Street View SnowMobile, a system used to capture on slopes.
-Street View Trike, used to reach some of the narrow alleyways that Google car cannot.
-Street View trekker , enables Street View to feature more places around the world — places no car, trike, trolley or snowmobile can access.
The photographs above may give you an insight into the conditions thrived to bring us closer to our world.
As part of the process, collecting imagery is just the starting. Then, matching each image to its geographic location on the map, by combining signals from sensors on the car that measure GPS, speed and direction. This helps to reconstruct the car’s exact route, and even tilt and realign images as needed. Turning into 360 degree photos and 3D images, is contributed by the rapid actions of three lasers above the car that reflect, detecting the distance between buildings and the position.This enables to build the model, by masking the images – giving a panoramic view.
Crucial information is extracted from the images such as business names, major vs minor roads, business logos, one way streets, street names and numbers, bicycle lanes – directly from the picture using computer vision algorithms.
Watch the space for part 2 – detailing more exciting features that allow us to navigate through the bustling cities, last minute family trips or just back home.
Video – Showcasing the use of Google Street View Trolley.