Page 70 - Security Today, May/June 2019

Register to attend free live events or view on demand after the live date! Click on the Webinars tab at securitytoday.com
Financially Secure – Top Trends in Banking Security
June 6, 2019. 2:00 PM ET
Sponsored by Salient Systems
Buying Some Time for First Responders
June 12, 2019. 2:00 PM ET
Sponsors include Johnson Controls, Salient Systems
On Demand webinars:
Video Surveillance — Understanding the Common Goal of Super Bowl LIII Sponsors include HID, Johnson Controls
Too Many Doors and Not Enough Control — How Video and Access Control Is
Key to Campus Security
Sponsored by Genetec
Why Schools Need a Mass Notification Unified Security System
Sponsored by Johnson Controls, Nightlock,
Safety Technology International, Salient Systems
Schedule your webinar today with
Kevin O’Grady, kogrady@1105media.com
Technology | Education | Solutions
Go to sp.hotims.com and enter 21 for product information.
36
Check Out Our Latest Webinars!
triggered based on face recognition or other analytics, even if that wasn’t the dedicated purpose of that specific camera.
Shorter development cycle. By develop- ing software that can be deployed on general purpose hardware—instead of developing the edge hardware itself—the end product is more broadly applicable, essentially shorten- ing the development cycle.
The Main Drivers of On-Camera Analytics
Today, due to memory and computation lim- itations, on-camera analytics tend to be used for point solutions. Initially, this was limited to motion detection, but on camera analytics have evolved and can now identify and clas- sify objects such as people and vehicles en- abling advanced activity such as intruder de- tection, license plate recognition and people counting. The big question in the VCA in- dustry today is whether edge devices will be- come sophisticated enough to enable general purpose identification, extraction, tracking and classification of all objects in the video. There are three main considerations driving development towards the edge.
Higher Demands for Real-Time Process- ing. Today, there is a higher volume of real- time data processing from a higher number of cameras. Because of these increasing de- mands, technology providers can justify the large initial investment in creating, market- ing and distributing smarter cameras to meet the demand.
Deep learning driving AI chip develop- ment. Now that deep learning is considered standard for video analytics enablement leading hardware providers are developing dedicated AI chips. Since these chips only support specific instructions required for deep learning inference, they feature high ef- ficiency, low energy consumption and small form factor.
Due to their flexibility, deep learning hardware solutions are enabling broad appli- cations. Autonomous cars, for instance, rely on this type of hardware, transplanting the deep learning enabling hardware in the car it- self, instead of in a centralized server center.
Lowering costs for decoding high-resolu- tion video. To run video analytics, captured video must be transmitted to recording ar- chives, live monitors or centralized process- ing servers, requiring significant bandwidth. By encoding video, solutions reduce trans- mission costs, but then face another obsta- cle: the work intensive demands of decoding higher resolution video, such as 4K.
A byproduct of processing video on the edge, circumventing video decoding ulti- mately reduces the computational require- ments for processing the overwhelming amounts of high-quality video data.
By the time the video captured by the edge device is transferred to the centralized location, it is already processed and can be decoded as needed. For post-event investi- gation, for instance, only the video for the relevant time and camera ranges need to be decoded. Thus, the extraction of evidence isn’t inhibited even though the demands of decoding have been reduced.
Balancing the Benefits of Edge and Centralized Computing
At the onset of 2019, VCA industry predic- tions focused on pain points driving the shift towards edge processing and cloud comput- ing—changes that will play a critical role in accelerating the adoption of advanced video content analytics. On camera analytics tech- nologies are focused on transforming from point solutions to offering a complete ana- lytics suite, including object tracking, clas- sification and recognition. However, for the foreseeable future, centralized computing will remain critical for deriving comprehen- sive intelligence from edge devices. To en- able cross-camera analytics, there must be a centralized computing service, aggregating insights from across cameras and feeds.
By overcoming decoding challenges, edge providers can drive enhanced operational reliability and processing speeds; reduced privacy risks by transmitting only encoded metadata; and, ultimately, accelerated migra- tion to cloud-based solutions. When comput- ing activities are limited to data and applica- tions and not video processing, centralized cloud platforms become a more affordable option for running inten-
sive video analytics, such as alerting, business intel- ligence, and video index- ing and search.
Tom Edlund is the CTO at BriefCam.
MAY/JUNE 2019 | SECURITY TODAY
ARTIFICIAL INTELLIGENCE
archy13/Shutterstock.com