How one city manages municipal services with AI

When Seongnam City deployed Alchera’s AI technology across its CCTV network, the city enhanced the quality of urban life without incurring significant administrative costs.

Challenge

In 2014, Seongnam City, South Korea (pop 950,000) established a unified administrative center for municipal services. The center aimed to improve the well-being of its citizens by collecting and analyzing real-time data related to traffic, the environment, public safety and facilities. Better monitoring, the argument goes, leads to better data which leads to superior quality of life outcomes. 

That’s the theory. 

With 305 cameras providing visual coverage of key roads, intersections, buildings, and public spaces, accessing a continuous stream of data was not a problem. Mindset was the challenge. Korean municipalities inching towards the Smart City concept focused, at that time, on public information and surveillance services, but not analysis. Relying on human judgment to consistently (and accurately) identify actionable information in a real-time video feed was, frankly, a Herculean task.

In the past, increased staffing might have been an acceptable solution. More eyeballs on CCTV monitors may have generated incremental improvements. In the age of big data, hiring more line workers does not always lead to more productivity. The sheer volume and velocity of data to process seemed to call out for a new mindset and big data technology.

That’s when Seongnam City called on Alchera.

Solution

Alchera transformed the city’s data analysis capabilities with advanced visual anomaly detection technology (VADT). In basic terms, VADT performs three core, simultaneous tasks:

  • read and digitize the CCTV video stream
  • identify and label people, vehicles, objects, and elements in the natural environment in time and space
  • automatically trigger alert protocols based on pre-programmed anomalies for specific locations and conditions 

VADT is a branch of artificial intelligence. It makes fast and accurate decisions. The veracity of those decisions depends on the quality of baseline data, the algorithm that labels video information, the parameters that define anomalies, and the system’s ability to learn new standards of normal and anomalous behavior. 

Anomalies are patterns of digital data that fall outside baseline parameters. A vehicle driving the wrong way on an exit ramp or a person jaywalking on the street are two such examples.

Results

Powered by Alchera, Seongnam City manages municipal services with smart technology. CCTV cameras provide the eyes and VADT provides the brains.

Here’s a shortlist of the technology’s features and benefits:

  • increase safety in public spaces through pedestrian and traffic monitoring
  • Decrease response time and cost for suspect search through human attribute search 
  • increase response times for accidents and natural disasters

The value proposition of VADT is enhanced safety and efficiency through accuracy and speed. Seongnam’s VADT system operates at processing speeds well beyond the human potential. VADT minimizes labor costs because there is no need for staff to continuously monitor the CCTV feed. Instead, staffing resources are allocated to high-value tasks such as inspecting alerts, validating anomalies, and monitoring solutions (e.g. tracking police and fire responses to road accidents or fires).

In other words, VADT allows staff to focus on events that improve the quality of urban life.

Quote

“The main constraint on innovation is no longer the difficulty in recording and storing information, but the finding of useful insights among the sheer abundance of data now being collected.” – Robert Fay and Wallace Trenholm

CCTV Facts

1942 first CCTV installed by Siemens to watch a V-2 rocket launch in Nazi Germany.

7500 number of CCTV cameras connected to an AI traffic system in Delhi, India

40% decrease in car wait times after installing 50 AI cameras in Pittsburg, USA

25% decrease in car travel time with the same cameras in Pittsburg

47% decrease in road wait times in Hagen, Germany after optimizing traffic light controls with Siemens AI cameras

About Alchera

We’re a technology startup from South Korea. Our domain is artificial intelligence (AI) and augmented reality (AR). Supported by a team of experts in computer vision, deep learning, graphics, data science, and software engineering, we solve problems. 

Since 2016, Alchera has empowered businesses to launch products, improve efficiencies, and bring the power of visual AI to their organization.

https://www.alcherainc.com

contact@alcherainc.com

Do you have any questions?

If so, feel free to get in touch with us!