You are currently viewing Traditional Edge systems

Traditional Edge systems

Traditional Edge systems

For simple applications such as factory monitoring and control system, various gateways are available to relay the data of sensors and PLCs to cloud. 

Typically, these gateways support MODBUS protocol to read data from devices. 

Following is the use case of a small factory with 3 machines 

There are two gateways which relay various parameters of these machines.

Gateway can relay data to cloud using MQTT or an HTTP request. There can be custom protocol sitting on top of TCP/IP also. Contineo platform supports all these protocols and can easily integrate these devices and gateways to the platform. Contineo provides REMO (Remote Monitoring) app for these purposes. These devices are fairly simple and may not have the ability to process data locally.

Another type of gateways are the wireless gateways, where sensors send data using plateau of RF interfaces like LoRA, Zigbee, m-BUS, BT-MESH etc. These gateway devices act like a mediator / data aggregator. Communication to server can be achieved using any of the protocols mentioned above.

Following is the use case for LoRA WAN devices

Next topic will discuss about AI powered devices which can be used for processing data locally using ML algorithms, also cover the video analytics using edge devices.

Advaced Edge Systems

EDGE AI –revolutionist in the field of video analytics and computer vision

Safety, security and surveillance are of paramount importance to any business that wants to protect its employees, customers, and assets from potential issues and malicious actors or optimise its functioning using video surveillance. Yet with more cameras and sensors than ever before, how can organizations lower their risk by quickly and efficiently analysing the flood of images and videos at their fingertips?

The solution comes in the form of edge AI for video analytics which can be applied to myriad of fields of which safety, security and surveillance are of paramount importance. We’ll showcase some of the computer vision and video analytics work together accomplished at our company and why it’s important to perform these analysis on the edge. 

  • Remote sensing: Many security cameras and sensors are located in remote areas, making it even more important to detect potential issues and risks. Computer vision can help identify suspicious vehicles, detect anomalies on water surface, and detect breaches of a boundary or perimeter, and much more.
  • Object recognition: Using cameras and State of the art deep learning algorithms you can more easily identify noteworthy objects and people within just a fraction of a second, rapidly determining if these objects and/or people pose a security risk.
  • Manufacturing process monitoring and defect detection: Computer vision applications have a major role in product and component assembly in the manufacturing space. As a part of industry 4.0 automation, most of the manufacturing industry has been implementing computer vision to conduct fully automated product assembly and management processes.

While computer vision has many possible applications, few of them are as time-sensitive as safety, security and monitoring. When a computer vision system identifies a potential anomaly/defect or a possible security risk for your organization, you need to take swift and decisive action which tells us time is of utmost importance when performing video analytics.

Unfortunately, many computer vision systems are too slow to perform real-time analysis: instead of processing the captured images or video themselves, they upload this data to a more powerful machine in the cloud. Latency issues (waiting for data to be uploaded and analysed) present a barrier to large-scale adoption of computer vision for safety and security AI.

That’s why edge AI is an essential development for the field of security video analytics. Edge computing is a paradigm in which data processing occurs in “edge” devices that are physically located close to the original point of capture, rather than by servers in the cloud.

CAM Agent

Contineo has an edge component called as “CAM-Agent” to facilitate, automate and integrate various image / video processing algorithms. The edge agent can connect to multiple cameras which are connected using USB interfaces or are a part of the local network. 

CAM Agent has built-in image processing algorithms and the CAM Agent can provide video/image data to custom algorithms that can be integrated with CAM agent with minimal amount of work. CAM agent also provides event processing service which can be used to push events to cloud and trigger video recording. The Contineo also provides analytics about various events using CAM Agent.

Future scope and areas of potential studies-

Distributed AI and Federate Deep Learning 

Distributed Artificial Intelligence (DAI) is a class of technologies and methods that span from swarm intelligence to multi-agent technologies and that basically concerns the development of distributed solutions for a specific problem.

It can prevalently be used for learning, reasoning, and planning, and it is one of the subsets of AI where simulation has a way greater importance than point-prediction. In this class of systems, autonomous learning processing agents (distributed at large scale and independent) reach conclusions or a semi-equilibrium through interaction and communication.

Federated Learning (FL)

Federated learning aims at training a machine learning algorithm, for instance deep neural networks, on multiple local datasets contained in local nodes without explicitly exchanging data samples. The general principle consists in training local models on local data samples and exchanging parameters between these local nodes at some frequency to generate a global model shared by all nodes.

Leave a Reply