Request Information

Machine Learning and IoT Video Analytics: Benefits for Smart Cities

Cities across the country have implemented video surveillance as a tactic to keep neighborhoods safer and monitor traffic and congested areas. Camera equipment has become cheaper, which means more municipalities can afford to install video surveillance in multiple locations around a city. And the benefits are clear. If an accident happens, authorities can review the footage to find out what exactly happened. If there is a crime in the region, again, authorities can review footage to find out valuable information. But what about all that footage that cities capture that is not relevant to anyone? Machine Learning and IoT Video Analytics have substantial benefits for smart cities.

machine learning video surveillance

Capturing and storing irrelevant information from video surveillance has become a problem. As Khamis Abulgubein, Principal Product Manager, IoT Applications for Nokia, notes:

Each camera produces 1MB of uplink traffic and storage per second, 24 hours a day, 365 days a year. Of that, only about 1% of is typically relevant data and even less contain data that requires an instant response, including traffic delays, streetlight outages, crime in progress, and other safety concerns. Compare those numbers with a city like Bristol, UK who has over 700 cameras, and you can extrapolate that it is a big problem.¹

He also noted the privacy concerns. Cameras continuously capture people’s movements on video surveillance footage, and that cities store data for an infinite amount of time, which is entirely unnecessary. Instead, what cities need to be able to do is capture the video, keep relevant footage, and delete the rest. Through machine learning, video surveillance can work smarter.

Solution: Nokia Scene Analytics

The current solution to the surveillance problem faced by municipalities is a third-party to capture and record the footage. However, this does not solve the problem of creating a quick response time to data anomalies (such as a car accident) or privacy concerns. The real solution lies with unassisted artificial intelligence (AI) capabilities that the Nokia Scene Analytics solution provides.

Nokia’s technology turns cameras into IoT sensors that help municipalities improve infrastructure and relieve citizens of privacy concerns. Instead of capturing data, storing it, and then only reviewing it if necessary, Nokia’s solution captures data, analyzes it in real time with AI, and alerts authorities to any anomalies. This ability means faster response times and reduced data storage since all unnecessary data can be deleted.

Nokia’s solution also means a human does not need to review all data, which is a manual and time-consuming process. Furthermore, unnecessary data will not be stored, which means fewer privacy concerns and less room needed for data storage. AI can point out ongoing anomalies such as congestion in specific streets, so authorities can take steps to fix the situation. The data can also help with future infrastructure projects.

Unassisted AI can revolutionize smart cities because, through machine learning, the technology will learn which data sets can be deleted and which need further investigation. This advanced technology can help smart cities improve everyday life for its citizens and better use data captured through surveillance.


References:

¹ReadWrite, Unassisted AI Video Surveillance Techniques Help Numerous Verticals to Scale (https://readwrite.com/2018/04/13/unassisted-ai-video-surveillance-techniques-help-numerous-verticals-to-scale/) , April 13, 2018.

Case Study: IoT is redefining the customer experience. Nokia case study.

© 2018 IoT Innovation All rights reserved.

Website by Ironpaper