An integrated software and hardware product for the edge intelligence scenarios of all industries, which deploys AsiaInfo’s rich and powerful AI capabilities to the edge in a lightweight manner. With variety of product forms to meet the user needs in various business scenarios, it can empower multiple industries such as security, energy, communications, etc., serve the society and improve people’s livelihood.
It has realized the transition from “post-passive investigation” to “active forewarning”, and from “watching video” to “using video”, with the characteristics of high capture rate, high accuracy rate, and low false alarm rate.
The monitoring screen displays alarms, and security personnel are actively informed through infrared, telephone and text messages to achieve a unified external interface; it can be integrated with existing equipment to achieve active processing of 5G IoT devices after an alarm occurs.
With task-based and pluggable security model encapsulation and configuration of composite security identification tasks, deep, multi-angle, and associated three-dimensional prevention and control can be realized.
Based on business scenarios, dozens of commonly used visual perception and cognitive models, such as fire recognition, vehicle recognition, and face recognition, can be preset, and online loading and updating of business models on demand, which is convenient and flexible.
The cloud carries large-capacity data storage and structural analysis, and the AIO provides fast and flexible AI reasoning capabilities for the edge. The collaboration of end, edge, and cloud can effectively cover the entire business process from monitoring, early warning to management.
Jumbo supports 320 channels of video stream access on a single device, and can concurrently provide 32 channels of AI algorithm services, with a recognition accuracy rate as high as 99.99%.
It integrates software and hardware for easy deployment, and is out of the box after simple configuration of plug-in, networking, one-key boot.
Flexible software and hardware interfaces support wired, WiFi, 4G/5G network access, providing two modes of web application and standard API to meet the needs of edge intelligence under different network environments.
The hardware body built based on the general platform can flexibly expand various modules such as AI acceleration, video processing, and network communication; the software layer is decoupled from the hardware to support unified computing and processing under heterogeneity, which can be flexibly configured on demand.