AI on the edge has increasingly become a trend. In a webinar held October 5, asmag.com discussed the various AI camera benefits with Hanwha Techwin, which also announced its new Wisenet X line of AI cameras.
AI on the edge has increasingly become a trend. In a webinar held October 5, asmag.com discussed the various AI camera benefits with
Hanwha Techwin, which also announced its new Wisenet X line of AI cameras during the program.
“AI and video analytics are ever evolving in the video surveillance market. Hanwha Techwin takes a leadership position in AI in the video surveillance industry. We have gained competitiveness in AI and developed relevant technologies,” said Eric Kim, Head of Sales for APAC at Hanwha Techwin, during his introductory remarks. “Hanwha Techwin has been enhancing its AI solutions line-up on a custom basis, aiming to develop complete and comprehensive AI solutions that satisfy customers’ needs.”
The webinar began with asmag.com unveiling its latest survey on technologies that impact security. According to the survey,
AI ranks on top both in terms of maturity and suitability. “Our survey points to a changing landscape in the security industry, shifting the focus to software development,” said Israel Gogol, Group Manager of asmag.com. “In the future, competition will no longer revolve around models in a manufacturer’s catalog, or their cost. Companies that will support better analytics and better tools and infrastructure for faster and reliable machine learning will have an advantage.”
The new Wisenet X AI cameras
Then, Elvin Boo, Head of Business Development at Hanwha Techwin SEA, announced their new Wisenet X AI cameras, which have the same feature sets as the company’s existing AI cameras including: object detection/false alarm reduction, whereby the camera can not only detect objects, but also discern objects separate from their surroundings such as waving trees, shadows, or animals; image optimization, whereby technologies such as WiseStream III adjust the video encoder to focus maximum video quality on the detected objects in a scene while scaling back encoding on the rest of the scene; and business intelligence, whereby the camera can detect and classify customers, for example, at any angle without additional cameras installation.
The Wisenet X AI cameras also feature the following that makes them particularly stand out:
Use of Wisenet7 SoC: The Wisenet X-core and X-plus cameras are equipped with Wisenet 7, Hanwha Techwin’s latest systems-on-chip. Performance-wise, the SoC enables the cameras to achieve 30fps at 4K and 120fps at 2MP, extreme WDR of up to 150dB for 2MP cameras capturing at 60 fps, and advanced lens distortion correction.
Cybersecurity: According to Boo, the cybersecurity of the new X AI cameras validates the boot process, securely stores key information, and preempts hacking risk with unauthorized access blocking. “Our cameras have been certified by UL-CAP for secure-by-default. The X-plus cameras have TPMs certified by FIPS 140-2, which is a security standard for cryptographic modules; the federal governments in the U.S., U.K. and Canada use the standard to approve whether a TPM is securely designed and implemented,” Boo said. “We can provide industry-leading end-to-end cybersecurity to the customer.”
Reinforced durability: The cameras also have reinforced durability features including: metal-shielded RJ-45 terminal and enhanced frame structures that protect the camera from transient voltage; hard-coated dome bubble providing better resistance to scratches; and power redundancy failover functions supplying DC/PoE power at the same time – continuous image capturing is possible without reboot even if one of the power sources is lost.
Resolution vs. analytics
The webinar went on to have a panel discussion on various topics concerning AI. The panelists agreed that aside from
resolution, onboard analytics have also become an important camera feature to look at.
“Resolution will still play a part, but it will not be the highest priority. I think analytics and other technology components of the camera are gaining traction among users, for whom AI can solve problems,” said Wenbin Sim, Sales Manager for SEA and South Korea at ICD Security Solutions. “AI brings the focus to the operator, instead of having the operator focus on multiple screens at the same time.”
“We do have queries coming in about what kind of analytics do you have on the camera and such. Resolution still plays a part – after all analytics is based on the field of the view of the camera. But resolution is no longer a priority question. It’s now more about what kind of analytics does the camera have and what does it do for the user,” Boo said.
Indeed, the things AI can do go beyond security and help users solve their various operational issues. “We're seeing some cases where analytics are used for hot-desking and space planning. We’re also seeing cases in retail; some even go to the extent of integrating with their POS systems. I think things like this will move along the way in terms of AI development in the future,” Sim said.
The panelist also agreed that further education is needed to ensure AI meets the user’s expectations. “With such technology advancement, we need to educate the operators, the users and even systems integrators on how you can actually operate AI and what is its limitation, so we can make full use of the technology,” Boo said.
Watch the webinar video recording: