With changing times, NVRs have become more than just recording devices amid more complex demands from users. Equipped with better hardware specifications as well as superb software and analytics capability, today’s NVRs are smarter than even and can help operators achieve various security and non-security objectives.
In a video surveillance system, the task of recording and storing IP video goes to the network video recorder or NVR. With changing times, NVRs have become more than just recording devices amid more complex demands from users. Equipped with better hardware specifications as well as superb software and analytics capability, today’s NVRs are smarter than even and can help operators achieve various security and non-security objectives.
Video recording has come a long way. Digital video recorders in analog systems have slowly given way to network video recorders amid a migration toward IP-based surveillance. And the growth of NVRs is not going to stop anytime soon. A recent report from Technavio forecasted that the growth of the NVR server market will grow steadily at a compound annual growth rate of close to 15 percent between 2017 and 2021.
In the beginning, NVRs recorded video footage mostly for post-event investigation purposes, providing evidence to end users, police and prosecutors after something happens. But increasingly, NVRs are equipped with more advanced functions and capabilities to meet more complex user demands. “The first key factor is ease-of-use, including easy setup and intuitive user interface. Then, users start to focus on further advanced functions such as H.265 HEVC to keep videos for longer time, 4K UHD display to easily check video in detail, integration of other systems/devices to centrally control and manage several devices and AI analysis and smart quick search when event happens,” said Jacky Cheng, CEO of
DIGIEVER.
“Traditional requirements for a network video recorder in a universal video surveillance system is that it should record videos of important moments and ensure that the video files are complete and not lost. With the global security situation going increasingly severe, people have gradually increased requirements and expectations for video surveillance systems. It is hoped that the key monitoring targets can be identified, the targets and abnormal behaviors can be discovered, and the operator can be alerted all in a real-time manner, so that timely measures can be taken to reduce the damage or loss caused by various abnormal conditions. In addition, after an abnormal situation occurs, relevant video files can be quickly called at any time, reducing the large amount of staff and time required by traditional solutions,” said Eason Rao, Product Manager at
Dahua Technology. “The development of artificial intelligence technology, efficient deep learning algorithms and powerful GPU computing capabilities enable intelligent NVRs to achieve real-time identification of targets or behaviors, and the speed of event queries is increased by tens of thousands of times. This not only greatly reduces the loss caused by a situation, but also the manpower and time spent on the incident inquiry.”
Smart NVRs
In this sense, NVRs have evolved into “smart” NVRs that fulfill the user’s various needs. “Most of normal NVRs will be replaced by smart NVRs, and they will be more and more economical and efficient in real-time object identification and metadata abstract,” said Rao. “The integration protocol will be unified as soon as possible, and it can work with third party devices and VMS.”
An enabler of smart NVRs is advances in hardware. Now, advanced CPUs and even GPUs have gradually become standard features in NVRs to process video from dozens of megapixel IP cameras. “Yes, more and more interests are based on structure data and object recognition, so powerful GPUs or CPUs with AI computing capability are more and more important,” said Rao.
A-MTK, meanwhile, has unveiled its NR3368S NVR that supports 720p to 4K IP camera resolution. “It’s based on ARM Cortex 4 core CPU. ARM-base means lower budget but good performance,” said Pauline Yen, Director of A-MTK.
The advances in hardware have in turn enabled more powerful software to be embedded in NVRs. Dahua, for example, has unveiled IVSS, its first AI NVR, an intelligent video surveillance server that combines video management functions with traditional video storage operations in one device. Besides the basic functions like video recording, live view and playback, IVSS is integrated with various intelligent features for different business and industrial applications. “These include facial recognition, human body detection, face features such as gender or whether the person is wearing glasses or mask, behavior analysis, vehicle recognition and vehicle features such as color, model and type,” said Rao.