This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies. Read our privacy policy>Search


To have a better experience, please upgrade your IE browser.

If you need help, please click here:

HUAWEI CONNECT 2018 — Activate Intelligence to Write a New Chapter of AI-Powered Security

HUAWEI CONNECT 2018 recently came to a successful close in Shanghai. Themed “Activate Intelligence to Write a New Chapter of AI-Powered Security,” Huawei Video Surveillance demonstrated its X series software-defined cameras and CloudIVS solution.

At the exhibition, Huawei presented its latest security solutions that focus on AI- and cloud-based smart cameras, intelligent analysis edge nodes, and Ascend chips with the highest computing power.

Huawei also demonstrated critical security technologies like all-weather scenario adaptation; personal and vehicle feature analysis and search; multi-camera collaboration; multi-algorithm integration; and all-cloud architecture-based video big data analytics and search.

HC2018 Video Surveilllance Booth

Software-Defined Cameras: Injecting New Vitality

Huawei believes that cameras in the future must provide on-demand scenario definition. Cameras must be able to load different software and algorithms according to different scenarios to quickly evolve through various combinations.

X-Series ITS cameras, face checkpoint cameras, vehicle micro-checkpoint cameras, omni-data structuring cameras, and situation awareness cameras all debuted at the exhibition.

Huawei software-defined cameras use the all-new Ascend series AI chips with a built-in Neural Processing Unit (NPU) and can perform AI algorithm training through deep learning technology, which helps identify more complex scenarios.

Cameras with Ascend chips can detect and track up to 300 human faces simultaneously. When compared with a traditional chip, an Ascend chip’s face detection is 20 times higher. Additionally, Huawei software-defined cameras can automatically optimize image parameter settings based on actual scenarios. With an Ascend chip’s computing power, a software-defined camera can connect to multiple nearby common cameras to give them intelligent analysis capabilities. A cluster of software-defined cameras can implement video analysis on edge nodes.

Huawei Software-Defined Cameras

CloudIVS: Device-Cloud Synergy, Matrix Intelligence

The Huawei CloudIVS Solution relies on AI and cloud technologies to implement resource sharing and efficient intelligent analysis, and build an open ecosystem.

The CloudIVS can be deployed at the center and edge nodes, and works with software-defined cameras to implement device-edge-cloud synergy and enable unified management, O&M, and application and multi-dimensional collaboration.

The Huawei CloudIVS 3000 is a typical configuration of the CloudIVS solution. It is specially designed to cope with video analysis service challenges, implement regional autonomy, and support fast service closures. The Huawei CloudIVS 3000 includes the following features:

Lightweight: CloudIVS 3000 requires only one server to get started. Its integration degree is 50 percent higher than the industry standard, and its overall power consumption is reduced by 30 percent.Speed: CloudIVS 3000 features computing on edge nodes, fast deployment, and a fast closed-loop for local services. Its algorithm integration duration is shortened from two months to two weeks, and system deployment is reduced from two weeks two days.Flexibility: CloudIVS 3000 features flexible video service loading. Its compute resources can be allocated on demand, which helps efficiently manage traffic bursts and globally coordinate tasks.Intelligence: CloudIVS 3000’s intelligent algorithm orchestrations precisely lock targets; feature on-demand scheduling for scenario-based data, and support practical applications.

The Huawei CloudIVS supports the concurrent operation of facial analysis algorithms from multiple vendors and can converge data based on the results of multiple facial analysis algorithms, implementing more precise face searches. Additionally, the CloudIVS allows users to quickly locate a target within a large number of images, even from behind the target image or with only a silhouette, improving investigation efficiency.