A new program that monitors street signs in need of replacement or repair, which utilises a technology that provides interactive panoramas along many streets in the world, was recently developed.
According to a recent press release, Geospatial scientists developed the fully-automated system is trained using AI-powered object detection to identify street signs in the freely available images.
The Problem
Large amounts of both time and money are currently being spent by authorities in order to monitor and record the geo-location of traffic infrastructure manually.
Councils have requirements to monitor this infrastructure but there is no cheap or efficient way to do so currently.
This manual task, in addition, exposes workers to unnecessary traffic risks.
With the use of free and open source tools, however, the team was able to develop a fully automated system for doing that job, and being able to do so more accurately.
The team found during investigations that mandatory GPS location data in existing street sign databases was often inaccurate, sometimes up to 10m off.
Tracking these signs manually by people, who may not be trained geoscientists, introduces human error into the database.
The AI system
Results of the project were just published in the journal of Computers, Environment and Urban Systems.
It showed that the system detects signs with near 96% accuracy. It can also identify the type with near 98% accuracy and can record the precise geo location from the 2D images.
The lead author of the study, a Geospatial Science Honours student from Australia’s RMIT University, explained that the proof-of-concept model was trained to see ‘stop’ and ‘give way’ signs.
Furthermore, it could be trained to identify many other inputs and was easily scalable for use by local governments and traffic authorities.
The system, additionally, can be used by any spatial analyst once set up. The user just needs to tell the system which area needs to be monitored and it would look after it.
Recognising the value of visual data
The project’s co-lead, who is a geospatial scientist from the University, shared that some councils were already attaching cameras onto rubbish trucks to gather street footage.
This shows how valuable visual data was becoming, given what technology could now do with it.
She added that imagery is critical for local governments in monitoring and managing assets. With the huge amount of geospatial applications flourishing, this information will only become more valuable.
Their project is one of several early applications for this to meet a specific industry need but a whole lot more will emerge in coming years.
Their system can also use footage from other sources, such as those from rubbish truck cameras or any other geo-referenced imagery of the road network collected by councils.
Their research can provide councils with an economical tool to drive insights and data from this existing resource.
