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/showpost/34324.aspx
INSIGHTS

AI-low light surveillance: top firms share insights

AI-low light surveillance: top firms share insights
Some of the leading video surveillance companies incorporate innovative strategies to take maximum advantage of low light cameras and algorithms.
Low-light surveillance solutions are becoming more and more sophisticated, thanks to recent advancements in hardware and software. Integrating artificial intelligence and machine learning has been the icing on the cake, enabling an incredible level of data capture.
 
Some of the leading video surveillance companies are at the forefront of this revolution, incorporating innovative strategies to take maximum advantage of cameras and algorithms. We recently spoke to Hikvision Digital Technology, Axis Communications, and Motorola Solutions to know more about their approach regarding this.
 
This article explores perspectives from these three companies and offers insights on how integrators can make use of what’s available in the market for better results.

AcuSense Technology from Hikvision

Hikvision has long demonstrated a deep engagement in the domain of low-light surveillance. Their proprietary AcuSense technology introduces a suite of intelligent features, empowering customers with a security solution that is not merely reactive but notably proactive.
 
“By integrating with AcuSense technology, our low-light imaging cameras can help achieve more security intelligence and effectively capture targets in true color,” said Rubio Hong, Product Director at Hikvision. “The AcuSense technology can distinguish persons and vehicles from other objects in the environment, minimizing false alarms triggered by animals, falling leaves, heavy rain, and other moving objects, which enables businesses and homeowners to focus security resources on real threats. While our low-light imaging cameras equipped with the Live Guard function can deter trespassers with visual and auditory warnings while sending notifications to users remotely.”

False alarm prevention is becoming increasingly important in today's security scene. With the digital transformation of numerous industries, the number of alert events has increased exponentially. As more technology is integrated into the operations of businesses, households, and public safety groups, the volume of data produced - and hence the possibility for false alarms - has increased.

False alarms can be more than an annoyance; they can also be a considerable waste of resources. Responding to a false alarm frequently diverts law enforcement or security teams away from real dangers, compromising overall security preparation. Furthermore, these circumstances can lead to 'alarm fatigue,' in which many false alarms lead to complacency, increasing the danger of missing a legitimate alarm.

Furthermore, false alarms might have significant financial consequences. Repeated false alarm activations may result in fines in many jurisdictions, while in a commercial setting, they can result in lost labor and operational inefficiency. As a result, false alarm prevention is a critical performance measure for modern surveillance solutions and plays a vital part in establishing the technology's return on investment.

Thermal and low-light performance with Axis Communications

Hiroshi Ochiai, Director of Alliances, Portfolio, and Customer Services in APAC for Axis Communications, said that one specific example here is the company’s edge-based intrusion detection analytics, AXIS Perimeter Defender with AI-based object classifications to be used with their thermal cameras.
 
“Using thermal cameras, you can visualize the scene even under complete darkness,” Ochiai said. “AXIS Perimeter Defender with AI-based functionality is based on deep learning algorithms, which can classify humans and vehicles without any user calibration. It provides optimal performance and greater reliability with fewer false alarms, even in challenging lighting and weather conditions. Another example is our edge-based object detection and classification analytics, AXIS Object Analytics.”
 
Utilizing AI-based algorithms, AXIS Object Analytics has the ability to detect, analyze, and classify various objects in a scene, including humans and vehicles. While this system is compatible with multiple cameras, it's crucial to note that for optimal utilization of these analytics, effective low-light performance is essential. Ensuring adequate results in low light is key to reaping the full benefits of AXIS Object Analytics and ensuring accurate, reliable surveillance results.

Leveraging convolutional neural network with Motorola Solutions

Hamish Dobson, Corporate Vice President for Product - Video Security and Access Control at Motorola Solutions, pointed out that Avigilon cameras are built with AI-powered video analytics to extend the effectiveness of security teams by proactively flagging potentially critical events in need of attention without the need for constant live monitoring by security teams.
 
“On the H5A Thermal camera, we have evolved our AI-powered analytics engine to a higher level with dedicated Convolutional Neural Network (CNN) training for the thermal imagers, improving object detection and classification accuracy,” Dobson said. “Previously, the analytics could only classify objects when they were moving, but now objects can also be classified even if they remain stationary (e.g., a parked car or a person standing still). Another significant benefit of the H5A thermal camera is its object detection and classification speed. The camera’s analytics can instantly detect and classify an object (person or vehicle) in real-time.”
 
In the past, analytics systems relied heavily on tracking an object's movement, taking several crucial seconds to identify its motion and shape before a classification could be issued. However, in the realm of perimeter security, this time lag can pose serious challenges.
 
For instance, a trespasser might move swiftly across a camera's field of view, potentially remaining undetected without comprehensive camera coverage. Therefore, speed in detection and classification is of utmost importance, and the advancements in today's technology are geared toward overcoming these latency issues for optimal security.

Conclusion

In short, the physical security industry is in a transformative period where technology is reshaping our approach to protection. Low-light surveillance solutions, backed by advanced algorithms and artificial intelligence, are becoming critical tools in our quest for safety and security.
 
Companies like Hikvision, Axis Communications, and Motorola Solutions are spearheading this revolution, offering increasingly sophisticated solutions that not only capture and analyze data but do so proactively, enhancing the effectiveness of security teams.
 
The strides made by such companies in AI-integrated surveillance solutions – whether it be Hikvision's AcuSense technology, Axis Communication's edge-based analytics, or Motorola's use of Convolutional Neural Networks – represent a seismic shift in the surveillance industry. These developments are enabling accurate object detection and classification, even under challenging conditions, and minimizing false alarms, allowing businesses and homeowners to focus on genuine threats.
Subscribe to Newsletter
Stay updated with the latest trends and technologies in physical security

Share to: