Sony AI-based image sensors are at the core of three smart city trials being run by Envision in Rome.

The Japanese company’s IMX500 image sensors with AI processing functionality could cut the city’s transport-related pollution and accidents at pedestrian crossings.

The trials’ primary objective is to evaluate and deliver a smart parking system to reduce pollution and gridlock from those cruising to find a parking space. For this, the trial seeks to evaluate the effectiveness of such a system, with drivers alerted via a smartphone app before being directed to the free parking space closest to the driver’s destination.

In addition, the tests include a study of smart city systems that will optimise capacity and increase the use of the public transport network by implementing smart bus shelters, counting those getting on and off each bus, identifying overloading to ensure better provisioning of buses and costs optimisation.

Finally, a system at pedestrian crossings will be progressively activated to alert drivers when pedestrians are crossing, using low-latency smart lighting on the road to make them more visible with the aim of reducing the city’s accidents on pedestrian crossings.

The trials are set to commence this month.

The image sensor configuration used for these trials allows the extraction of real-time metadata related to information of a free parking space, the presence of a pedestrian about to cross a street, or the number of people getting on or off a bus. No images are stored, nor leave the sensor, in line with privacy requirements.

This allows the city to reduce the bandwidth required, to scale it using existing networks, and to cut power and communication costs.

Traffic caused by those seeking to park is seen by the city as a key contributor to congestion, with the trial aiming to quantify the scale of the problem.

The average distance between pedestrians and vehicles is a key metric used to measure pedestrian safety. This trial aims to deliver a quantitative analysis and prevent pedestrian accidents through signalling mechanisms installed at the crossings.

“This is a clear and concrete example where Sony’s smart vision can serve our customer’s purpose while respecting people’s privacy,” said Antonio Avitabile, managing director of corporate alliance and investment at Sony. “We have a vision of achieving more sustainable and liveable cities and, through the IMX500 scalable platform, we can substantially accelerate this process.”

Genius version smart tips have been installed in Rome’s city centre. A preliminary phase was undertaken to train a neural network to identify available parking spaces as well as the number of people waiting at the bus stops, entering and leaving the bus, and waiting to cross or crossing the road.

Every Genius smart tip consists of two sensors looking over the roads around and the parking spaces. The sensors send real-time data elaborated by neural networks on the exact location of a free space, the pedestrians’ presence and the number of people queuing at bus shelters.

The exact location of the free parking space data is streamed in real time through the smart tip. The data are then immediately processed by the sensor integrated in the smart tip, using neural networks, and sent to Envision’s cloud software. The coordinates of the free parking space’s location are overlaid in real time on a map that is displayed on a mobile device used by the driver who is heading towards the area.

Pedestrians’ presence is measured and compared across different locations. The neural network system detects pedestrians at zebra crossings and a lighting signal is sent to the drivers to alert them.

Data of queue length and people getting on and off the bus are processed by the sensor in the smart tip through the neural network and sent to the Envision software platform which aggregates them and make them available to personnel managing the public bus network to enhance the planning and scheduling of the bus transportation network. A crowded figure of merit is provided to signal when the bus is running at over capacity to avoid overcrowded buses, better manage the transportation network and improve citizens’ journey experience.

The trial is conducted in the city of Rome collaboratively by a number of start-up companies in the Italian ecosystem and with the support of Sony:

• TTM Group (DP Control) is responsible for installing the IMX500 image sensor in the smart tip.
• Envision is responsible for the development of the smart tips. Envision designs and develops end-to-end infrastructures for smart cities and IoT, managed through the UrbanVision software platform.
• Citelum is a public lighting company with experience in the areas of road signs, video surveillance, urban electric vehicles and services, and wifi networks. It is the promoter of the project and responsible for the installation of the smart tips on traffic lights, and manager of the maintenance of traffic lights as well as of luminous signs and traffic regulation and control in the city of Rome.