Join or Sign in

Register for your free asmag.com membership or if you are already a member,
sign in using your preferred method below.

To check your latest product inquiries, manage newsletter preference, update personal / company profile, or download member-exclusive reports, log in to your account now!
Login asmag.comMember Registration
https://www.asmag.com/project/micron_edge_storage_for_video_security/
INSIGHTS

Audio in video surveillance: An overview

Audio in video surveillance: An overview
When audio technology is added to video surveillance, it can complement video and boost an organization’s security posture significantly. This article looks at the value and benefits of integrating audio with video surveillance.
Video surveillance plays a central role in security. When audio technology is added to video surveillance, it can complement video and boost an organization’s security posture significantly. This article looks at the value and benefits of integrating audio with video surveillance.
 
Needless to say, video surveillance only “sees” a scene and alerts operators when suspicious activities are detected. Yet increasingly, users are realizing the benefits of audio, which can complement video surveillance to contribute to a more effective and responsive security strategy.
 
“Audio reveals unique insights beyond what a camera can see. When integrated properly, audio enhances video surveillance by providing important context and improved situational awareness. AI-enabled cameras with onboard audio analytics can detect high-risk sounds, such as glass breaking, yelling, or gunshots, immediately trigger alerts, even before an intruder appears on screen. This capability allows operators to verify incidents faster, reduce false positives, and respond more proactively,” said Adam Lowenstein, Product Director at i-PRO Americas.
 
“Combined with video, audio capture has the potential to reinforce the great majority of existing surveillance use cases. For example, security operators can get a significantly better overview of scene events if their video stream is complemented with an audio stream,” said Carl Malmqvist, Regional Director, South Asia Pacific, and Dinesh Rai, Business Development Manager for Network Audio, Asia Pacific, at Axis Communications. “Just as you may employ several types of video analytics for automatic event detection and alarming based on visual detection, audio analytics can monitor the audio streams and react when something stands out.”
 
Beyond sound capture, audio can also complement video surveillance by serving as a deterrent: When someone suspicious appear in a scene, the operator can speak live or play a pre-recorded message to the individual through an adjacent speaker to scare the person off.
 
“For live surveillance operations, today’s solutions offer organizations the option to use both pre-recorded messages and live, operator-initiated audio via remote loudspeakers,” said Sean Um, GM and Head of European Business Development Team at IDIS. “Pre-recorded announcements can be automated, manually activated or scheduled to provide routine reminders. Direct, in-person communications can allow operator intervention in the event of a threat, such as emergencies or security incidents.”
 

What types of audio can be analyzed

 
Today’s audio solutions in cameras can identify various sounds that may signal distress or trouble, thanks to advanced technologies such as AI. The audio can be analyzed on the edge instead of being sent to the backend for analysis.
 
“Modern AI-capable cameras can detect and classify a variety of security-relevant sounds, including glass breaking, screaming, gunshots, vehicle horns, and persistent alarms. These sounds are processed directly on the edge using powerful AI chips like the Ambarella CV-52 used in i-PRO cameras. There’s no need to send audio to a central server. This edge-based approach means faster alerts, reduced bandwidth usage, and, most importantly, enhanced privacy,” Lowenstein said.
 
“Axis cameras that are compatible with AXIS Audio Analytics – AI-powered edge analytics – detect important sounds, such as breaking glass, screams, shouts, and speech (only that someone is speaking, not actual words). AXIS Audio Analytics can also measure SPL (Sound Pressure Level), which is the intensity of sound measured in decibel (dB),” Malmqvist and Rai said. “All in all, analytics can be trained to ‘learn’ specific sound events, categorize them, and recognize them for future events. If AI-based algorithms are used, they can be trained from large amounts of data. For example, an algorithm can reliably detect human screams after having been trained with thousands of such sounds.”
 
According to IDIS’s Um, while utilizing audio feeds to monitor live sounds from various locations will enable operators to detect unusual activities or disturbances, a common-sense approach is required, depending on the location of each site.
 
“Familiar sounds, such as human voices and shouting near cameras in a town or city center, are often to be expected in busy urban areas and do not require a security response. However, continuously monitoring audio gives operators the ability to quickly visually verify if such sounds may indicate anti-social behavior or criminal activity, particularly at night,” Um said.
 
“Similarly,” he adds, “dogs barking in built-up areas during darkness may be the result of disturbance by passersby. However, in a more remote location, a facility, or where guard dogs are used, the same sounds may signal trouble, and surveillance operators will need to verify whether there is a potential threat. The same is true for sirens or other alarms, as well as vehicle sounds such as revving or screeching tires, which might be heard before cameras pick up suspicious behavior. By contrast, the sound of shattering glass will often indicate a break-in or vandalism that requires a rapid response.”


Product Adopted:
Other
Subscribe to Newsletter
Stay updated with the latest trends and technologies in physical security

Share to: