Hikvision releases new traffic cameras and servers

Hikvision has announced the release of a suite of checkpoint cameras, traffic incident detection cameras and servers.

The company says that the new cameras and servers are powered by advanced large-scale Traffic AI models.

Enhancing traffic safety

According to the company, these innovative products are designed to significantly enhance detection accuracy, reduce false alarms and improve overall traffic safety.

AI models are used for smarter detection with Hikvision’s large-scale traffic AI models based on the Guanlan Large-Scale AI Model platform.

The cameras and servers have been developed with deep traffic industry knowledge and are trained on millions of complex real-world situations such as:

• Rainy roads
• Construction areas
• Extreme weather

The models use the Transformer architecture, an AI-building approach that allows simultaneous viewing of data parts.

It assists the models by identifying complex features and can rapidly determine the relationships between different elements in a traffic scene.

Hikvision explains that the self-attention mechanism in these models enables better extraction of overall features.

Instead of just focusing on individual details, the cameras analyse the whole picture to understand the context, which will significantly reduce false detections caused by the background.

Seatbelt detection

According to Hikvision, the self-attention mechanism evaluates the driver’s posture and visible parts of the seatbelt, even when partially obscured and reduces false alarms by 75%.

The models pre-training generalisation and self-attention mechanisms allows them to work well in complex situations, such as:

• When the seatbelt color matches the clothing
• The belt has an unusual shape

Detecting phone use

When it comes to detecting phone use, Hikvision says that the cameras analyse multiple factors such as:

• Hand position
• The area near the ear
• The driver’s line of sight

By integrating industry knowledge rather than relying solely on the phone’s shape, they reduce false phone use detections by 75%, meeting the need for high accuracy.

Traffic incident cameras and servers

60% lower false detection rates Hikvision’s incident detection cameras (iDS-TCS802-HS, iDS-TCS907-HIR, iDS-TCS917-HI, iDS-TCS900-HI) and servers (iDS-TSS500-H/16, iDS-TSS300-H/04) leverage large-scale AI models to deliver a 60% reduction in false alarm rates for incidents like fallen objects, pedestrian detection and stopped vehicles, the company explains.

The cameras minimise background false detection by better understanding the image context.

Hikvision says that their enhanced generalisation ability allows them to detect rare targets and capture complex features, such as:

• Identifying event types based on key characteristics of personnel and vehicles

For example, they can identify reflective vests or emergency lights to accurately classify events.

The servers let users get these benefits using their existing cameras, which avoids the need to install new ones for road monitoring.

“A big step forward”

Hollis Li, Intelligent Traffic Product Director, Hikvision commented: “These new cameras are a big step forward in traffic management efficiency.

“With large-scale traffic AI models, our solutions are more accurate and better at handling real-world challenges.”

Li concluded: “This means safer roads and more efficient traffic management for everyone.”