Intelligent Digital Video Processing System for on-site Monitoring

Location : Japan

Background Information

Video Processing System is applied in numerous domains, like manufacturing, traffic monitoring, medical inspection, and long-distances education. Nowadays, thousands of cameras are used in real-time monitoring and video analyzing to ensure the situation on site. Now the booming applications for video processing are from transportation sector, we use traffic monitoring and people counting to improve the efficiency in our city. If an emergency occurs, operators in the station can identify incidents immediately through cameras so that they can take action quickly.

Digital Video Processing System utilizes real-time video technology to detect incidents automatically from local. However, traditional video systems suffer from low video quality which results in inaccurate analysis. By using Intelligent Digital Video Processing System, traditional video system application can be much improved. Combine with Artificial Intelligence trained models, local situation can be easily detected and tracked, and the situation can be easily controlled from central management team.

System Requirements

It is necessary to have multiple cameras covered the sites needed. Each camera conducts the situation in its own monitoring area to detect abnormal events. The system needed to handle several cameras and be able to detect the dangerous or unwanted signs by using AI deep learning and training methods.

Apart from the detection system, all cameras are monitored by inference system in the center control room. If an incident occurs at the roadside, in intersections or railway the staff in the control room have to be able to control the situation spontaneously. Therefore, in the surveillance applications, it also needs back-end systems to trigger alarm messages and to notify staffs.

To make the whole application functioning, related systems must be built up, like railway signaling, and warning alarms in traffic monitoring and security system, so that the system can make awareness in the incident. This Intelligent Digital Video Processing System is expected to reduce the burden on employees and the cost of time and effort for incident detection.

System Description

Video detection problems frequently occur in traditional video analytics system. However, AI video detectors benefit from deep learning technology, they not only detect the abnormal scene but also recognize objects precisely. HPC-8212 is the edge server to grab video streams from IP cameras and recognize objects based on AI inference. HPC-8212 can work smoothly with SAS transmission 12Gb/s and 4 NVMe drives supported via an optional riser card to provide greatly effect of video processing.

In central control room, the AI inference server monitors all cameras in the different fields. To process all this massive amount of information, HPC-8212 provide big storage capacity with 12 3.5” HDD for massive amount of information saving. HPC-8212 features twelve 3.5” 10TB SATA HDD, an AI storage technology from Seagate. The Seagate SkyHawk AI drive is designed for DVR and NVR systems. 

Key Features

• High performance computing AI server with high transmission speed
• Industrial AI systems with big storage capacity to perform AI inference
• Deep learning computing at the backend permits an intelligent management

Why Advantech

To fulfill real-time video monitoring in the different fields, Advantech offered an AI solution that meet their application scenario. The rich product offering of AI systems makes it easy for customers to develop their edge solution. HPC-8212 2U rackmount server needed to support high speed transmission and big storage capacity all in a single server.

Related Product