Archive News

November, 2018



Solution Transforms Cameras into Active Vision Systems Capable of Event Detection and Analytics

TORONTO – NOVEMBER 14, 2018 – Arcturus Networks Inc. demonstrated today an intelligent vision analytics system capable of detecting various public safety, security and operational events from within a stream of live video. The demonstration illustrated the application of advanced machine vision algorithms combined with Convolutional Neural Network (CNN) object classification methods to identify events such as motion, intrusion, boundary crossing, zone incursion, loitering, motionless behaviour, abandoned package, face verification, removed object, crowd estimation, crowd flow and counter flow. The demonstrations addressed two key factors with existing video surveillance technology –the reliance on human monitoring and the escalating demand for bandwidth as camera deployments grow and resolutions increase.

Real-time Analysis

The demonstration involved continuously processing video in real-time by applying a mix of background subtraction, tracking and object classification methods to identify different events and behaviours. When an event occurred, such as a person crossing a virtual boundary from one zone into another zone, the system produced an incident notification. The notification gets used to alert personnel and streamlines their workflow by providing them a complete video sequence of the incident. This level of automation reduces the continuous performance workload of analysts and improves their situational awareness, resulting in faster, more effective responses.

“Existing camera systems are passive, they require events to be observed and reported before an analyst gets involved to review the footage, then determine the situation,” said David Steele, director of innovation at Arcturus. “By using machine vision, we can transform cameras into active detection systems capable of identifying events, such as a person falling from the edge of a subway platform onto the tracks. Response time is critical in this situation and analysts need to understand how the event transpired to determine if it was accidental or criminal.”

Reduced Network Traffic

Existing analytics systems rely extensively on central “big-iron” infrastructure and cloud services. This approach requires the transmission of pixel data from each camera to a central system for processing. As the proliferation of surveillance cameras escalates and resolutions increase this places upward pressure on the bandwidth capacity of the network, resulting in data bottlenecks, extensibility problems and the requirement for new infrastructure to support capacity expansion. Edge-based analytics provides a distributed approach where video is pre-processed at the edge of the network resulting in small-packet event notifications and video-on-demand, instead of every pixel of data. This approach eliminates the need for ever-increasing bandwidth capacity and makes video surveillance a suitable application for systems connected using a pay-per-bit providers, such as public transit vehicles connected via LTE.

Pricing and Availability

To request a demo or for more information contact Arcturus sales.

More Information

Videos from the demonstration include Boundary Crossing/Line Incursion and Abandoned Package Detection.
Additional videos can be viewed at the Arcturus YouTube channel.
Preliminary information on the vision product offering can be obtained from the product landing page. Vision Middleware product landing page.
For more information or to set up a demo, contact Arcturus.

About Arcturus

Arcturus is a leading provider of embedded software and hardware for the connected world. Arcturus enables companies globally by delivering specialized, full-stack solutions for voice, video, vision and IoT, with an emphasis on trust-chain security and system integration. Arcturus has been developing products for OEMs since 2002 and applies its focus to security, surveillance and communications applications in public transportation, healthcare and building system markets. For more information, visit

Press Contacts:
Arcturus Networks Inc.
416.621.0125 x233

# # #