Unlocking the full potential of edge-AI cameras

March 28, 2023

Artificial intelligence has revolutionised video security by providing unprecedented control over situations. Security cameras that were once reactive devices used for investigative purposes can now offer real-time alerts, allowing users to take proactive action against threats.

Market analysts expect the penetration of AI in physical security to increase in the coming years. According to OMDIA’s 2023 Trends to Watch: Physical Security Technologies report, implementing AI applications will remain the key market focus this year.

“In the field of video surveillance, the global market size for AI chips in security cameras will reach $1.3 billion in 2026, according to the latest data from Omdia’s forecast,” the report said. “Current estimates suggest that AI cameras will account for 42 percent of all network cameras shipped globally in 2026.”

The need for edge AI

High performance video analytics used to be server-based because they required more processing power and storage than a camera could offer. But algorithm development and increasing processing power of edge devices in recent years have made it possible to run advanced AI-based video analytics on the edge.

There are benefits to using edge-based video analytics applications – while AI applications can run on the server or cloud, latency and bandwidth concerns limit their performance. In situations where an access control solution must provide instant results after analysing a face or a number plate, sending data to the server and waiting for its answer is not feasible.

This is why edge analytics is becoming an area of significant investment for video security systems. It offers low bandwidth consumption as only necessary data is sent to the server or cloud. Limiting processes to the edge ensures quicker alerts in case of threat detection, allowing faster data-driven analysis and decision-making. Edge-based analytics also come with lower hardware and deployment costs as fewer on-premises server resources are needed for the security solution.

This however brings unique challenges. Edge analytics applications need to work on edge devices that can have memory and processing power constraints. Seamless management is critical to smooth operations.

Challenges when using cameras with AI on edge

AI enabled cameras provide more value than their traditional counterparts. But they also demand more resources in terms of data and hardware.

New video security cameras are now being designed to process multiple video streams and higher resolutions (4K and above) to provide AI algorithms with the large dataset of detailed images and videos required for them to analyse. In addition to this, increasing amounts of metadata are captured and stored on the device to enable operators to quickly search and find relevant video footage. Much of the processing now occurs on the device level, with more computing power from new chipsets enabling deep neural network processing on the camera itself for edge intelligence. Memory and storage technologies then need to keep up with these evolving changes in processing and workload requirements.

“With edge devices generating critical insights for everything from public safety to vehicle autonomy to manufacturing operations, today’s smart applications cannot afford to compromise on latency or quality,” said David Henderson, Director of Industrial Segment, Micron Embedded Business Unit. “Micron’s high-performance, ruggedised solutions — our i400 microSD card and 1α node based DRAM — will unlock new value for businesses and drive the rapid innovation needed at the intelligent edge.”

Get reliable performance for a variety of deployment scenarios with Micron’s memory and storage products – delivering high capabilities and industrial temperature options in several small form factors.

Learn more about Micron