WHAT IS AN EDGE AI APPLIANCE?
An edge AI appliance refers to a hardware device that integrates artificial intelligence (AI) and machine learning algorithms to perform tasks and make decisions at the edge of a network, rather than relying on a centralized server.
This allows for real-time processing and analysis of data close to its source, reducing latency and bandwidth requirements. Edge AI appliances are commonly used in industries such as IoT, industrial automation, and smart cities to enable devices to make intelligent decisions locally without the need for a cloud connection.
Vision AI is used in a variety of applications, including but not limited to: self-driving cars, surveillance systems, retail and advertising, and medical imaging.
OUR EDGE AI APPLIANCE: CLICK
CliCK is the edge AI solution developed by Eurotech and WaterView to extract from cameras real time information on extreme weather events and their effects.
CliCK generates data and alerts that may raise people’s awareness, reduce risks and losses, save lives, thus supporting climate change adaptation strategies to increase community resilience.
WHY DEPLOY AI AT THE EDGE?
WHAT ARE THE BENEFITS OF CLICK?
Edge AI produces actionable insight where it’s needed, by using video analytics, machine learning, artificial intelligence and sensor fusion on suitable hardware and electronics.
One billion cameras capture 30B images a second (100 trillion images an hour) throughout the world, and many of them contain weather related information. CliCK processes huge amounts of video data to identify risky weather situations quickly and automatically, making our cities smarter and safer while offering typical edge AI benefits such as:
Data transfer to cloud and back takes time. Plus, in extreme weather, you cannot take remote connections for granted
Analyses taking place in a fraction of a second are crucial: in critical weather, the faster the better
Reduced data transmission and data stored in the cloud
Balance between what makes sense to keep at cloud level and what can be stored and processed locally
IDC estimates there will be 55.7 billion connected IoT devices by 2025, generating almost 80B zettabytes (ZB) of data. As volumes grow, centralized services and data transfer become a true bottleneck, that can only be avoided by locally processing data
To guarantee a true compliance with privacy regulations (like EU GDPR) processing sensitive data locally and in real-time is key because it only exists for a blink of an eye before it disappears. Less sensitive data stored in the cloud also means less risks of cyberattacks
HOW DOES CLICK WORK?
CliCK is a one-of-a-kind solution embedding advanced AI weather analytics on a cybersecure edge appliance powered by NVIDIA GPUs, with a complete set of tools for remote device/app management, Video Management System (VMS) integration, local/remote data exchange
- IEC 62443-4-1/2 cybersecurity certifications
- Integrated anonymizer for faces and license plates
- Azure IoT and AWS IoT certified
- Embedded Everyware Software Framework (Java/OSGi) IoT framework (ESF) and Eurotech Everyware Cloud (EC) native enhancing secure device management and remote access
- weahterCAM: to detect rainfall and snowfall events and estimate their severity
- floodCAM: to identify flooded areas on the ground
- snowCAM: to detect snow
- visCAM: to estimate visibility distance in presence of fog, smoke and precipitation
- smoCAM: to detect smoke plumes and fire in outdoor
* Up to 40 instances overall (one per camera)
- IP cameras
- Minumum resolution 2 Mpx
- Sensor size > 1/2.8″
- Max focal length 10mm
- Support for RTSP and http protocols
- Embedded IR illuminator (for rainfall and snowfall detection)
- External sources of visible light for flood, snow cover and visibility detection in nightime
Although CliCK is camera agnostic and can easily integrate with multiple on-prem and cloud third party solutions, it is natively compatible with products of trusted best in class partners.
EXAMPLES OF CLICK USE CASES
- Heavy rain and snowfall detection
- Flood detection along rivers and waterfronts
- Smoke plumes and wildfire detection in protected areas
- Smoke detection in oil & gas plants, or in waste recycling sites
- Fluid spill detection in industrial plants
- Visibility distance estimation
- Flood detection in sensitive areas like underpasses
- Snow settling detection on road pavement