Google Street powers AI tool to monitor road infra

June 20, 2019:

A study shows the system detects signs with near 96 % accuracy, identifies their type with near 98 % accuracy & can record their precise geo-location from the 2D images.

The fully-automated system is based on AI-powered object detection to identify street signs in the freely available images.

Published in the journal of Computers, Environment & Urban Systems, the study shows the system detects signs with near 96 % accuracy, identifies their type with near 98 % accuracy & can record their precise geo-location from the 2D images.

“The proof-of-concept model was trained to see ‘stop’ & ‘give way’ (yield) signs, but could be trained to identify many other inputs & was easily scalable for use by local governments & traffic authorities,” said the study lead author Andrew Campbell from RMIT University in Australia.

Municipal authorities spend a large amount of time & money monitoring & recording the geo-location of traffic infrastructure manually, a task which also exposes workers to unnecessary traffic risks.

“By using free & open source tools, we’ve developed a fully automated system for doing that job, & doing it more accurately,” Campbell said.

Source Link