AI cameras to detect violence on Sydney trains – Software

CCTV cameras on Sydney’s significant rail network will be augmented with synthetic intelligence above the subsequent six months to automatically detect and report suspicious and violent incidents.

Transportation for NSW options to trial the technological innovation to analyse footage captured by the cameras, as section of a new initiative to enhance security for women of all ages travelling on general public transport at night.

It is just 1 of 4 winning suggestions from the Security Just after Dim Innovation Challenge, which supplied candidates fairness-free of charge seed funding and guidance by means of TfNSW’s digital accelerator.

Scientists from the College of Wollongong’s Smart Infrastructure Facility pitched the AI computer software, which can automatically analyse actual-time camera feeds and inform operators.

“The AI will be experienced to detect incidents this sort of as folks battling, a group of agitated individuals, folks adhering to a person else, and arguments or other abnormal conduct,” Smart lecturer and team guide Johan Barthelemy reported.

“It can also recognize an unsafe ecosystem, this sort of as wherever there is a absence of lights.The program will then inform a human operator who can rapidly react if there is an problem.”

PhD student Yan Qian, who will be aiding Barthelemy on the project, reported open up source code will be employed to “estimate the poses of a human getting and predict if there’s a fight”. 

“The incident will then be reviewed by a human controller who will acknowledge or reject the suggestion designed by the AI,” Qian reported.

“In this way, the software will become smarter, discovering in a very similar way to a human getting.

“As much as we know, absolutely nothing like this has been attempted globally. We are pushing the boundaries of the technological innovation.”

Other winning projects consist of a project pitched by data sharing system She’s a Group to use datasets and algorithms to develop general public transport routing that prioritises security.

A new system for general public security and aid, pitched by security technological innovation experts Guardian LifeStream and Cardno/UNSW, will also be trialled. 

Transportation Minister Andrew Constance reported the “winners were being preferred for their possible to meaningfully handle actual security issues” making use of “creative and advanced new technologies”.

“We want all our buyers to experience harmless on the network and it is not great enough that nine out of 10 Australian women of all ages expertise harassment on the street and modify their conduct in response,” he reported.

“We’re excited to be functioning with entrepreneurs and universities to implement ground breaking technological innovation remedies to continue to keep women of all ages harmless.”