Page 4 of 3

More Than a Set of Eyes

• Feb 01, 2019

The technology behind today’s optical sensors and lenses has come a long way in trying to mimic some of the hallmark characteristics of the human visual system. When it comes to video surveillance, the human brain can limit the effectiveness of even the best camera system. Our brains have an amazing ability to filter out information even when we don’t realize it. Similar to digital compression, our brains tend to ignore redundant data, particularly when other distractions are present.

Tune In, Tune Out

There are studies that illustrate how security staff, when looking at a video monitor with no action, will tune it out after just 20 minutes and completely miss an event when it finally does take place. Multiply that by the number of cameras and monitor windows and the problem only compounds itself.

Any surveillance system where humans are part of the monitoring process is subject to this type of human error. It is no one’s fault, it is just the way we are made, and may actually allow us to better process information based on an internal “priority.” While it might be possible to throw more bodies at the problem, it wouldn’t really fix the underlying issue, so we can still predict that events will be missed.

Cameras don’t miss a thing of course, and with motion tags written to an NVR, it is easy to go back and quickly find incidents after they have occurred.

What about stopping events in progress? Security personnel can have a lot on their minds. Depending on the size and scope of an organization, they may be required to multi-task, doing much more than simply staring at a monitor waiting for something to happen. What operators really need is help to detect events as they happen, in real time, so that attention can be focused when seconds count.

When a Camera is More Than a Camera

IP-based cameras today can do so much more than their analog counterparts and “seeing” is only a small part of it. Analytics are not new to our industry, but using them for focused, real-time response to events is still underused. Even for customers who own cameras with analytics features, many of them never get the follow-through with their integrators to set them up correctly. This is an opportunity to add real value to the system, but typically the analytics are only used to set motion markers or tags on captured footage for easier post-event searching. There’s so much more that cameras can do when it comes to alerting staff when an event is happening.

Take the example of a business that closes in the evening and should have zero motion occurring inside the building after hours. If an intruder somehow manages to bypass the standard security system by tricking a door sensor, should a camera not be able to send an alert that motion is occurring when it should not be? These simple additions are often overlooked by integrators, installers and customers alike.

Audio Analytics Enable Focused and Quick Responses

When glass is smashed, and a business is in the process of being burglarized, it is important to be notified immediately. If a camera with a microphone can detect and correctly identify the sound of glass breaking, then it is a perfect complement to any security system and can help reduce overall costs of installing purpose-built glass break sensors at every point of ingress.

Gunshots, screams, or even just noise going beyond a normal threshold can all be detected by modern on-edge camera analytics, yet few of the myriad number of IP-based surveillance systems deployed make making use of this technology.

Audio analytics can quickly pinpoint zones that security staff should focus on, which can dramatically shorten response times to incidents. Audio-derived data also provides a secondary layer of verification that an event is taking place, which can help prioritize responses from police and emergency personnel.

Is it Legal to Capture Audio?

With many state laws governing audio recording, audio analytics on the edge overcomes legal challenges as it never passes audio outside of the camera. The result of audio analytics processed at the camera is simply an event message saying a certain type of sound was identified.

Processing audio analytics in-camera provides excellent privacy since audio data is analyzed internally with a set of algorithms that only compares and assess the audio content. Processing audio analytics on the edge also reduces latency compared with any system that needs to send the raw audio to an on-premise or cloud server for analysis.

How Do In-Camera Audio Analytics Work?

Many IP-based cameras have small microphones embedded in the housing while some have a jack for connecting external microphones to the camera. Microphones on indoor cameras work well since the housing allows for a small hole to permit sound waves to reach the microphone. Outdoor cameras that are IP66 certified against water and dust ingress will typically have less sensitivity since the microphone is not exposed. In cases like these, an outdoor microphone, strategically placed, can significantly improve outdoor analytic accuracy.

There are several companies that make excellent directional microphones for outdoor use, some of which can also combat wind noise. Any high-quality external microphone should easily outperform a camera’s internal microphone in terms of analytic accuracy, so it is worth considering in outdoor areas where audio information gathering is deemed most important.

The camera extracts the characteristics of the audio source collected using the camera’s internal or externally connected microphone and calculates its likelihood based on the pre-defined database of audio signatures. If a match is found for a known sound—gunshot, explosion, glass break, or scream—an event is triggered, and the message is passed to the VMS.

Surveillance cameras with a dedicated System on Chip (SoC) have become available in recent years with in-built video and audio analytics that can detect and classify audio events and send alerts to staff and emergency staff. Having a SoC allows a manufacturer to reserve space for specialized features such as an audio analytics database of reference sounds needed for comparison or room for specialized business analytics.

Analytics for Business Intelligence

In the retail space, business analytics applications enable people counting, heat maps, and queue management. Similar to audio analytics, no records are kept about personal identities, only actionable insights about what is working and what needs improvement. This capability is seeing rapid adoption as more integrators and end users discover the consistent business value inherent in what was previously a capital expense. A security system that delivers actionable intelligence about businesses and city infrastructure can rapidly pay for itself and even become a revenue generating tool in some instances.

Just as computers and VMS continue to get “smarter,” so too are cameras, evolving their capabilities to see beyond the pixels they capture. IP-based cameras can do much more than simply pass images to a monitor or NVR. Modern IP-based cameras with sophisticated system-on-a-chip designs are completely self-contained security devices capable of detecting a range of events and behaviors both visually and aurally.

Hanwha Techwin’s Wisenet cameras include a suite of built-in analytics. We continue to invest heavily in R&D to bring new levels of edge-based analytic capability to our customers as technology and deep learning algorithms continue to evolve. There also is focus on cyber-security and installation flexibility has allowed us to develop a unique SoC that both protects and enhances our cameras in a way that is truly industry leading. In-camera audio analytics are a unique Wisenet camera feature.

As advances in AI and machine learning continue to evolve on-edge analytics, it is time to set expectations higher for security systems and infrastructure. Make sure your camera system goes beyond simply providing video feeds and becomes a key component to a holistic security infrastructure.

When considering a new system or updates to an existing one, look for manufacturers with a solid track record and a constantly evolving product line. Most importantly, think about what else your security system can do for your business with regards to analytics, not just for security, but also for business intelligence. You might just find it pays for itself.

This article originally appeared in the January/February 2019 issue of Security Today.