Autor: SENSING FEELING LIMITED
Smart visual sensing for real world spaces for safety and risk management. Ready for Private 5G.
Sensing Feeling delivers smart visual sensing software for real-world spaces, powered by Computer Vision and Machine Learning.
SensorMAX™ Edge - Ready for Azure Private MEC & Private 5G Core
Our Computer Vision technology is revolutionising the way our clients analyse and operate their real world spaces. By using advanced software algorithms known as Deep Neural Networks, our products enable clients to exploit imaging assets such as cameras to interpret and understand visual data just like humans do to help design and operate these spaces for better management of end-user safety, wellbeing and security.
With the integration of IoT and 5G technology, we are now able to take this to the next level by creating smart, connected devices that can process and respond to visual information in real-time.
One of the key benefits of using our Private 5G MEC solution is the ability to process complex data closer to the sensor sources, reducing latency and increasing the speed of data transfer, whilst providing the security and privacy of an on-premises solution. This is especially important for Computer Vision applications as it allows for faster and completely private processing of visual data.
Our Computer Vision IoT solution is designed to take full advantage of the capabilities of Azure Private MEC and Private 5G Core deployments and are specifically designed to be used to better manage safety, security and risk in transport infrastructures, industrial workplaces and smart cities.
What is a Private 5G MEC?
A private 5G Mobile Edge Computing (MEC) is a type of network infrastructure that allows for the processing of data closer to the source of the data, rather than in a centralised data centre. This is particularly useful for Internet of Things (IoT) and Computer Vision applications as it can help to preserve privacy by reducing the amount of data that needs to be transferred to external cloud destinations, or in deployments where no external Internet connectivity is available, or prohibited for security reasons. In a typical IoT application, data from sensors or cameras is transmitted to a central location for processing. However, with a private 5G MEC, this data can be processed at the edge of the network, closer to the sensors generating the data. This enhances performance and adds security whilst preserving privacy.
- At construction sites for upholding safe systems of work & enforcement of people-plant interfaces
- At venues and events for the safe management of movements, crowding & clustering behaviours
- In shared-use spaces for improving safety and amenity through analytics of traffic and people flows
- In safety-critical energy & maritime locations for improving worker safety with automatic detection of human risks
- In busy and complex spaces for improving safety and risk management using existing CCTV assets
- In manufacturing and industrial supply chains for the management of worker safety & wellbeing
- Real time 5G IoT edge processing
Our product exploits 5G connectivity by taking advantage of high-speed, low-latency communications between imaging sources (cameras) and our edge processing node. It utilises Computer Vision powered by AI Machine Learning algorithms to process and analyse the visual data streamed from cameras deployed within the Private 5G network, enabling them to extract real time insights to help our clients to make decisions or take actions based on the data points or alerts generated.
- Complete on-premises solution
- Our Private 5G-ready software solution is delivered as an on-premises package designed to be installed and operated within an organisation’s own physical facilities, rather than being hosted in a remote, cloud-based data centre. The product is specially designed to take advantage of Private 5G Mobile Edge Computing (MEC) technologies, which offer scalable and trusted environments for ‘in-network’ computing to support our powerful Computer Vision processing pipelines.
- 'Pick and mix' algorithms
Our product is powered by an ever-expanding range of dynamically-selectable ‘out of the box’ Computer Vision algorithms designed to support a variety of use cases. Simply ‘pick and mix’ the detection algorithms to suit your needs, whenever you need them.
- Custom vision models
For applications that are very unique to your context we can train and deploy new Computer Vision models to solve the problem dramatically faster and at much lower cost than traditional methods.