New from Hanwha Vision is its second-generation X series AI cameras. Six models are available, powered by the new Wisenet 9 AI-native System on Chip (SoC).
Designed for diverse settings, the developers say that the cameras offer clearer images, smarter monitoring, and more efficient data handling for security operations – while delivering AI object detection and attribute extraction, as well as enabling control over AI functions. At the core of the second-generation X series AI cameras is Wisenet 9, with dual neural processing units. While one NPU handles image processing, the other focuses on object detection and advanced analytics. This dual-NPU concept was introduced to ensure video quality and analytics have independent resources, for preventing one function from impacting the other.
AI
The AI event detection pipeline is separated into three independently adjustable streams: metadata streaming, intelligent video analytics, and image enhancement. This allows users to tailor detection settings, utilise specific features, and reduce unnecessary alerts. Intelligent video analytics on the cameras use AI object detection to trigger events. The X series AI camera features analytics such as vehicle counting, people counting, and queue management to address a range of customer needs. Analytics such as slip and fall detection, face mask detection, and social distancing detection are available.
The range features BestShot, a function that captures and transmits the most relevant images of classified objects. These images, cropped around the detected objects by the AI, accelerate forensic searches. Metadata such as gender, vehicle type, and clothing colour is transmitted alongside the images to further streamline the search process.
With AI-based Wide Dynamic Range (WDR), the models capture details in dark and brightly back-lit areas by reducing saturation in bright spots and improving shadow visibility. AI Noise Reduction minimises noise in low-light settings, for sharper images, the manufacturer says. For privacy protection, AI-powered Dynamic Privacy Masking (DPM) provides real-time safeguarding of personal information at the edge with objects, such as people, faces, vehicles and licence plates, dynamically masked with a mosaic or colour overlay.
RE-ID
The range also includes RE-ID; a deep learning-based approach to matching visual identities across non-overlapping camera views. For example, it can identify a person entering a building with an orange jacket and later recognise the same person leaving with the jacket unzipped and a hat, across different cameras. As for cybersecurity, with Secure Storage, Secure OS, and Secure Boot, these cameras prevent the execution of unauthorised software and ensure the integrity of captured data. The cameras are FIPS 140-3 Level 3 certified. Features of the X series AI range include:
• Powered by the Wisenet 9 AI-native SoC
• Independent AI control over metadata streaming, analytics, and image enhancement
• Dynamic Privacy Masking (DPM) to protect personal information at the edge
• bandwidth reduction with up to 50 per cent saving
• AI-based WDR and noise reduction
• Enhanced cybersecurity – FIPS 140-3 Level 3 certification
• Hardware tampering detection.
Visit HanwhaVision.eu/.
