Sony involved in a Smart City project in Rome 

Sony's IMX500 image sensor trial aims to reduce traffic congestion, optimise public transport and improve pedestrian safety through the use of intelligent vision sensors

90
Sony coliseum Smart City

Sony Europe has announced that IMX500 image sensors with AI-based processing are at the heart of three smart city trials that will be carried out together with Citelum and Envision in the city of Rome. The IMX500 is an image sensor with Artificial Intelligence processing functionality embedded in the sensor itself, addressing privacy concerns, and can be deployed using existing infrastructure.

The main aim of the trial is to evaluate and implement a smart parking system based on the IMX500 sensors, which will reduce pollution and traffic jams resulting from the need to drive around for a long time looking for a parking space. The system will send a notification to a smartphone via an app, indicating the nearest free parking space to the driver's destination.

The test is also aimed at studying smart city systems that optimise the capacity of the public transport network and increase its use, through the adoption of smart shelters at bus stops, capable of counting passengers getting on and off each vehicle, so as to detect overloads and ensure better service planning and cost optimisation.

Finally, an alert system is to be gradually activated at pedestrian crossings to signal motorists when pedestrians are crossing by means of a smart floor lighting system to make them more visible, with the aim of reducing accidents at pedestrian crossings.

The sensor configuration used for the experiment allows the extraction, in real time, of metadata derived from information about a free parking space, the presence of a pedestrian about to cross the road or the number of passengers getting on or off a public transport vehicle. The images are not saved or transmitted by the sensor, in compliance with privacy laws.

The city can therefore reduce the bandwidth needed to transmit data, even using existing networks, and thus cut energy and communication costs.

Antonio Avitabile, Managing Director of Corporate Alliance and Investment at Sony comments: "This is a clear and concrete example of how Sony's smart vision solution can meet customer needs while respecting people's privacy .Our ambition is to make cities more sustainable and liveable and, through the scalable IMX500 platform, we can accelerate this process substantially.

The technologies behind the experimentation

The Genius smart tips were installed in the centre of Rome. In the preliminary phase, a neural network was trained to identify free parking spaces, the number of passengers waiting at bus stops and getting on and off vehicles, and pedestrians in the vicinity of crossings.

Each smart Genius tip consists of two sensors facing the surrounding streets and parking spaces. The sensors send, in real time, data processed by neural networks regarding the exact position of the identified free parking space, the presence of pedestrians and the number of passengers waiting at the bus stop shelters.

Data on the exact location of free parking spaces is transmitted in real time through the smart tip. The data is then immediately processed by the sensor inside the tip, using neural networks, and then sent to the cloud-based software platform of the experiment partner, Envision. The coordinates of the available parking space are displayed in real time on a map on the display of the mobile device used by the driver to drive to the marked area.

The presence of pedestrians is measured and compared at different points in the city. The neural network system detects pedestrians crossing the zebra crossing and sends a light signal to motorists to signal them to cross.

The data, processed by the sensor inside the tip using neural networks and relating to the number of people waiting at the bus stop and boarding and alighting from the bus, is sent to an Envision software platform, which aggregates it and makes it usable for public transport network managers to optimise transport planning and scheduling. Based on a figure of merit calculated as a 'crowding' parameter, the system can signal when a bus has reached its maximum capacity to avoid overcrowding, better manage the transport network and improve the travel experience for citizens.


You might also be interested in:

Smart City: are we on the right track for sustainability?

 

Previous articlecongatec expands with the birth of congatec Korea
Next articleLeonardo and Ericsson: 5G agreement for industry and public safety

LEAVE A COMMENT

Please enter your comment!
Please enter your name here