Article sourced from: Dell EMC Solutions Subscription Account

There is a job like this

Basically does nothing on ordinary days

Just sitting on a high chair

Looking straight ahead

No chatting, no sleeping, no playing with phones

Even going to the bathroom has to wait until the shift change

And this state often has to be maintained for several hours

Sounds boring, right?

This is the daily life of a lifeguard, a job that seems boring but carries the greatest responsibility in the world—saving lives. Lifeguards need to remain highly focused, scanning the area like radar, observing potential drowning situations and swiftly providing rescue.

According to a report by the World Health Organization, over 300,000 people drown each year worldwide. In many places, lifeguards scan their designated areas using only their eyes or binoculars, a traditional method that raises widespread concern—water is both alluring and dangerous, especially in open water where undercurrents are strong and situations are complex. Can lifeguards without technology capture danger in time and rescue drowning people?

The AI startup Sightbit is determined to address this concern with technology. The founder graduated from Ben-Gurion University of the Negev in Israel (the university known for developing strange communication methods to steal data), and they have developed a danger detection system deployed at beaches to assist lifeguards in their rescue work.

This danger detection system is based on convolutional neural networks and image detection models, providing lifeguards with views of potential dangers. The system trains on tens of thousands of beach images using NVIDIA GPUs, and its deep learning and proprietary algorithms can individually identify children and crowds, marking outliers.

During the pilot phase, Sightbit collaborated with the Israel Nature and Parks Authority to install three cameras, feeding camera data into a single edge AI supercomputer at the local beach lifeguard tower, providing real-time beach alerts, tracking, statistics, and risk analysis instant reasoning.

With this danger detection system, lifeguards can receive real-time updates on ocean conditions and identify dangerous rip currents offshore, warning people to stay away from these areas. Thus, when lifeguards start their shifts, they can be aware of the dangers present that day and prevent accidents.

This is a typical application of the combination of edge computing and artificial intelligence. As a complement to the cloud, edge computing brings computing, storage, and networking closer to the source of data creation, enabling faster data processing, decision-making, and insights. Coupled with the widespread availability of high-performance AI processing capabilities, AI has become the primary workload for edge deployments.

In many cases, data is most valuable at the moment it is generated, and immediate action is often required—for example, detecting a drowning person at the beach, inspecting defects on an industrial assembly line, or identifying obstacles on the railway.

In situations where every second counts, sending data to a data center is not ideal; for AI applications that require real-time rapid responses, moving AI inference to the edge is key to advancing business development.

The combination of edge and artificial intelligence will bring significant advantages in many industries:

Healthcare, where AI and edge servers can accelerate image analysis for pathologists while reducing human errors;

Retail, where the combination of video analytics and computer vision makes it feasible to gather insights into customer behavior, facilitating targeted marketing and loss prevention in retail environments;

Manufacturing, where edge analytics can monitor the quality of manufacturing processes and provide alerts when issues arise on the production line.

So how can edge AI inference be achieved?

As a leader in the field of artificial intelligence, Dell Technologies drives innovation in its products, building infrastructure to run AI workloads and providing significant value for organizations’ digital transformation with deep expertise and a broad product portfolio.

Dell Technologies provides comprehensive integrated solutions for edge AI, computing, storage, and networking solutions, including:

● Integrated Solutions— Proven AI designs, including computing, networking, storage, software, and services optimized for AI workloads

● Computing— From laptops and workstations to servers and HCI solutions

● Storage— Edge data requires enterprise-grade, scalable storage and management

● Networking— A complete set of networking solutions and switches for integrating edge devices into existing enterprise networks

● Hyperconvergence— VxRail satellite nodes extend automation and lifecycle management capabilities to the smallest configurations of edge workloads to date

Additionally, Dell Technologies works closely with NVIDIA to provide a wide range of NVIDIA-certified systems suitable for various use cases from the core of the data center to the network edge.

For many organizations, the answer to edge AI lies in the NVIDIA inference platform:

This software provides critical performance, efficiency, and responsiveness across cloud, data center, network edge, and automated devices to support the next generation of AI products and services.

The NVIDIA inference platform provides low latency and high throughput for inference applications, unlocking the full potential of NVIDIA GPUs in Dell PowerEdge servers, which have been certified by NVIDIA, allowing enterprises to confidently deploy solutions to run their AI-accelerated workloads securely and optimally.

Organizations ready to embrace edge IT need both strategy and partners with experience in AI and edge. Dell Technologies has the expertise, products, services, and influence to simplify the edge with inherent security and provide insights where they are needed most.