Google continues to develop its Project Sunroof service, which allows to evaluate the efficiency and feasibility of installing solar panels on the roofs of houses in a specific locality (taking into account the roof slope, the height of the sun above the horizon, the number of sunny days per year, the presence of shading objects).
Now the service has a new tool – Data Explorer, showing a map of existing solar installations in nearby areas throughout the US. Data Explorer combines the technology of computer learning with the analysis of images from Google Maps and Google Earth.
The developed algorithms of machine learning can now automatically find and identify solar installations in photographs. This can be both photovoltaic panels that produce electricity, and solar hot water heaters.
Project Sunroof, launched by Google in 2015, has already covered 50 US states and has become available for the first time outside the country. Now it is launched in Germany.
Function Data Explorer while working only in the US, where he analyzes data on nearly 60 million buildings. At the moment, Data Explorer has discovered about 700 thousand solar installations in the US (although the Association of Solar Energy Producers (SEIA) says about 1.3 million units).
According to Senior Software Engineer Project Sunroof Karl Elkin, the main task of Data Explorer is to help people make informed decisions about whether to invest in solar energy.
Google believes that the invention will stimulate people to use solar energy