Recently, IDC, an international authoritative market research institute, released the "China Edge Cloud Research, 2021" report for the first time, and Sophon Edge, the edge computing platform of Star Ring Technology, was selected as a typical product of the edge cloud software platform in the edge cloud solution market.
Sophon Edge is a marginal, intelligent cloud-edge converged computing platform that addresses the process of multimodal data integration and governance. Sophon Edge supports standard video and IoT protocol access, low-code business processing, high-performance data processing and analytics, enterprise-grade cloud-edge data, service governance, and adaptation to heterogeneous hardware such as edge embedded and cloud servers.
Typical of edge cloud
Based on the definition of edge cloud, IDC studied the main product forms, market size, and typical application scenarios of edge cloud, and made predictions on market development trends and potential opportunities. Although China's edge cloud market is still in the early stage of development, under the current limited application scenarios, the edge cloud market still shows a good growth trend in terms of commercialization.
The Star Ring Edge Computing Platform has been widely used in cloud-edge integrated scenarios
Over the years, Star Ring Technology has continuously improved the ability of Sophon Edge to perceive intelligence, and has been widely used in cloud-side integration scenarios such as intelligent manufacturing, intelligent security, intelligent construction sites, intelligent transportation, smart cities, smart campuses, intelligent gas stations, and equipment predictability maintenance.
Sophon Edge product architecture
Sophon Edge industry applications
Taking steel manufacturing enterprises as an example, in the steelmaking process, there needs to be a measurement of temperature and carbon content as a reference for the control end point. After the relevant equipment parameters and process data of the steel production line are collected, the edge node Edge Node is responsible for the basic equipment management and gateway services, and the device messages after docking are connected to the industrial Internet platform through The Flip, the self-developed real-time stream computing engine Slipstream of Star Ring Technology, and the production model training is carried out in the cloud Edge Hub, and sent to the edge platform for real-time application. Continuous application is achieved by grayscale upgrade and automatic upgrade of the model at the edge end.
Based on The Edge Computing and real-time stream engine technology of Star Ring Technology, it realizes real-time monitoring and alarm of on-site process execution to ensure that the production process is carried out in an orderly manner. At the same time, the temperature and carbon content hit rate of the steel output point are improved, and the quality of molten steel is 19 to 33% higher than that of experienced production. After the model is issued, the production parameters can be remotely controlled in real time, which can increase the production efficiency of the steelmaking process by 5%, reduce the intervention of labor on the terminal control, and reduce the total working hours invested by personnel for 45 days per month.
In the smart supervision scene, taking the illegal dumping of sewage by street catering enterprises as an example, in the past, law enforcement personnel needed to collect evidence on the spot for illegal acts such as sewage dumping of shops along the street in order to punish, and this manual supervision method undoubtedly increased the difficulty of urban management, and it is urgent to change from manual supervision to intelligent monitoring. Based on the edge computing platform Sophon Edge, the self-developed illegal dumping AI algorithm intelligently analyzes the video, realizes the real-time identification of illegal dumping behavior, and returns the location information of the illegal behavior and the gif dynamic picture of the illegal behavior back to the urban operation data center. The 24-hour uninterrupted monitoring and identification of illegal acts of sewage dumping has realized the collection of evidence for illegal acts, allowed law enforcement personnel to have evidence to rely on and evidence to check, effectively reduced the manpower input of urban management, and ensured the quality of urban water.
The Sophon Edge platform can also inject AI computing power into the edge, process the data nearby, reduce the need to transfer data back and forth between the clouds, efficiently process the results of face data, face biometrics, crowd analysis, etc., empower personnel supervision, prevent personnel from over-the-wall intrusion, escape from boundaries, illegally invade dangerous areas and other abnormal behaviors, and ensure the property and personal safety of enterprises.
Taking the construction site scene as an example, the construction site is often a variety of work and mixed personnel, often unable to carry out effective and timely supervision; production accidents and casualties also occur almost every day; not only that, dust and noise have always been the main problems in disputes between the construction site and nearby residents. In view of the above problems, Sophon Edge, the edge computing platform of Star Ring Technology, can complete the following work, personnel basic information management: personnel attendance information statistics, real-time number of people at the construction site; automatic equipment ledger management: equipment operation status monitoring, equipment failure prediction analysis, mixer working status monitoring, muck truck entry and exit information statistics; violation and dangerous behavior early warning: personnel gathering detection early warning, worker cross-boundary identification, worker wearing non-compliant alarm, hoisting detection; Sophon Edge's recognition capture rate for personnel is greater than 99%, the recognition rate is more than 95%, and the recognition delay is milliseconds; since its launch, more than 3,000 violations have been corrected, ensuring zero safety production accidents at the construction site.
In the process of construction of the urban rail intelligent video analysis system, the previous subway detection method is still manual inspection, which requires a lot of manpower every day, and the overall work efficiency is not high, and it is urgent to have a video intelligent analysis application to intelligently analyze the equipment monitoring video, intelligently identify abnormal states, and improve the inspection efficiency. The Sophon Edge Computing Platform provides a one-stop real-time anomaly detection platform based on deep learning algorithms such as efficient and appropriate image classification, object detection, and instance segmentation, which can continuously and iteratively update the model to improve the accuracy of defect detection in the production line and adapt to business changes.
The video-based bow network spark detection system developed by sophon Edge, an edge computing platform, detects the ignition event during the operation of the entire line, conducts statistical analysis of the number of sparks and the size of the sparks, provides a guiding basis for catenary maintenance, and evaluates the health status of the subway line according to the frequency and location of the bow network sparks, and establishes a complete set of arc quality health evaluation system. Compared with relying on traditional manual monitoring methods, the detection system based on video recognition significantly saves labor costs.
With the deep integration of a new generation of information technology and the industry, digital transformation has increasingly become a new driving force for enterprise development. Artificial intelligence is becoming the core of enterprise digital transformation by reconstructing enterprise business processes, user experience, product services and business models. Previously, Sophon Edge, the edge computing platform of StarHub Technology, participated in the "EC Ready" edge service test initiated by the China Academy of Information and Communications Technology and passed a comprehensive and rigorous evaluation. StarLink technology can provide more open data access capabilities, provide more comprehensive data governance capabilities, smoother one-click deployment capabilities and easier-to-use process configuration capabilities, and strive to become an advanced edge computing platform in the industry.
The new infrastructure is a new driving force for economic recovery in the post-epidemic era, closely related to IT spending such as artificial intelligence, 5G, and the Internet of Things, and the combination of edge computing and these emerging technologies will further promote the development of the market. In the future, as more edge scenarios emerge and users gradually realize the potential value of edge computing for business innovation and process optimization, edge cloud may become a new growth engine in the entire IT solution market.
IDC Edge Cloud Research Area