Artificial Intelligence, as a key part of any smart city initiative, can help keep people safe. Yanbu has successfully shown how AI manages some pressing issues amid urbanisation. As one of the biggest cities in Saudi Arabia, Yanbu has applied Huaweis Intelligent Traffic Management system to reduce road fatalities, with the ultimate goal to change the behaviour of road users and thus becoming a smarter, safer and better city.

As urbanisation progresses, innovative technologies are applied to help build or rebuild cities. Governments and administrations are thus looking for more diverse ways to manage increasing complex issues related to peoples livelihood. Meanwhile, the number of vehicles and how well the road systems are developed have become major indicators to reflect the level of urbanisation.

While such developed cities are getting busier, so are their traffic. The World Health Organisation has estimated that about 1.35 million people are killed on roadways all around the world each year. It means an average of 3,700 road fatalities were recorded every day. More than half of these are pedestrians, motorcyclists, and cyclists. In advanced cities, the number is even higher as traffic control is much more complicated than we could imagine.

Saudi Arabia is a forerunner in embracing digital innovations and new technologies. In 2016, the country announced the Vision 2030 programme, aiming to transform into a smart nation with a series of goals to achieve in three key stages. As one of the major industrial cities of the country, Yanbu Industrial City has taken up the same vision and begun its transformation.

Yanbu Industrial City, commonly known as Yanbu in short, is a major Red Sea port city in the Al Madinah Province of western Saudi Arabia. It has a population of some 240,000 in an area of roughly 606 square kilometres. Many of them are foreign expatriates working in the oil refineries and petrochemical industry.

Though most of them are from Asia, there are still large numbers from the Middle East, Europe, and North America. Known for its industry activities, especially when it is the worlds third largest refining centre, the road traffic is usually busy.

Saudi Arabia is a forerunner in embracing digital innovations and new technologies

Like all other stories of advanced development, Yanbu began with a limited number of surveillance cameras at some major interactions. The devices couldnt do more than just capturing still images, not to mention any functions like video recording, zooming, or some sophisticated recognition and search capability.

The set-up was simply not enough to catch up with the rapid changes stemmed from enlarging population and more frequent road commutation. And with limited surveillance, people just didnt care much about breaking traffic laws putting road users in a more dangerous situation.

This is where artificial intelligence (AI) can play a big role. Collaborating with Huawei, Yanbu installed the e-Police system in 2019. The eyes of the system are indeed the 256 high-definition cameras working at 16 major intersections, providing numerous quality images and videos, which allow authority to trace vehicles or impose different controlling measures to keep the whole road network smooth and safe.

The upgraded hardware in the software defined cameras (SDCs) can feed data into the system to further action. The track-and-trace and the occasional real-time controls require more than just sophisticated programming gimmicks.

Not only are the cameras come with improved image quality and 20 TOPs computing power, but the systems are optimised for extreme environments with intelligent light compensation mechanisms.

Pattern recognition is not new to AI, so the system is already capable to run video search based on license plate number and structured vehicle data such as makes and colours, looking to find out the vehicles of interest.

The Intelligent Traffic Management System (ITMS) puts AI into a more intelligent practice by running smart algorithms to automatically identify traffic violations, such as running red lights, crossing lanes, reverse driving, and lane marking infractions, etc. The query response is much faster, running about 100,000 data records and getting a response within seconds.

The precision is also stunning at 50-centimetre lane-level track recognition, with target and tracking recognition rates both over 95 per cent per cent. Traffic incidents can be detected in just eight seconds, with a monitoring distance of 200 metres, covering the major parts of the scene. Violation snapshots and information are then transferred to the central platform for unified storage and query, while the system automatically processes violation cases.

Continued here:

With AI, Yanbu has become smarter and safer - SmartCitiesWorld