Alex Bell likes to bike round New York Metropolis, however he bought fed up with how typically bike lanes have been blocked by supply vans and idling automobiles. So he determined to do one thing about it, the New York Instances stories. Bell is a pc scientist and he developed a machine studying algorithm that may examine visitors digital camera footage and calculate how typically bike and bus lanes are blocked by different automobiles. He skilled the algorithm with round 2,000 pictures of several types of automobiles and for bus lanes, he set the system to have the ability to inform the distinction between buses which are allowed to idle at bus stops and different automobiles that are not. Then, he utilized his algorithm to 10 days of publicly accessible video from a visitors digital camera in Harlem.
It took Bell round three weeks to develop the algorithm and his system took a few day to research the visitors cam footage. The outcomes confirmed that the bus lane on the block coated by the digital camera was blocked 57 p.c of the time whereas the 2 bike lanes have been blocked 40 p.c of the time. Primarily based on these numbers, Bell decided that roughly 850 automobiles had blocked the bike lane throughout these 10 days and 1,000 blocked the bus lane.
And whereas these findings are primarily based on only one metropolis block and a brief time frame, the opposite 101 miles of bus lanes and 435 miles of motorcycle lanes all through the town endure from the identical points. It is an issue many individuals desire a resolution to. In that regard, Bell has made his supply code accessible on GitHub within the hopes that extra folks will use it and get a extra correct understanding of the scope of the issue. “We have now large issues particularly on the subject of transportation,” Bell instructed the New York Instances. “I’m hoping that simply seeing the information will assist it’s much less ignorable.”
By way of: New York Instances