The Intelligent Perception Gateway is a purpose-built edge AI visual gateway designed for smart cities, intelligent transportation, industrial monitoring, and security applications. It integrates a high-performance AI chip, advanced video processing modules, and proprietary deep learning algorithms. This device can process high-definition video streams in real time at the edge, enabling intelligent functions such as object detection, behavior recognition, and anomaly alerts, helping users achieve efficient and precise monitoring and decision-making in data-intensive environments.
Key Features
Multi-Protocol Support
Compatible with HTTP, RTSP, RTMP, RTP, Bonjour, and more, enabling seamless integration with almost all video terminal devices on the market and flexible connectivity with third-party platforms.
Wide Range of AI Algorithms
Supports over 40 intelligent algorithm models including facial recognition, intrusion detection, post abandonment detection, people counting, crowd flow statistics, specific behavior analysis, dress code recognition, flame and smoke detection, vehicle and license plate recognition, and more.
Edge AI Processing
All recognition tasks are performed on-device, significantly reducing the need to transmit video data to the cloud. This accelerates response times and achieves millisecond-level real-time detection, providing a solid foundation for proactive safety alerts.
Structured Data Output
AI recognition results are output in structured data format, which can be linked with video surveillance data to enable automatic anomaly alerts, statistical analysis, and further big data processing — empowering decision-makers to stay informed and act promptly.
Integrated Video Management
Supports multiple high-definition video stream inputs, enabling video surveillance resource sharing across devices and platforms. Compatible with various protocol-based platforms.
Self-Learning and Model Updating
The system supports online learning and periodic algorithm updates, continually optimizing recognition accuracy across different tasks and scenarios, reducing false alarms, and achieving ongoing intelligent evolution.
Remote Technical Support
Compatible with remote maintenance platforms, allowing technical support teams to offer prompt remote services to users.
Hardware Specifications
(*Base configuration; customizable per deployment*)
* CPU: 8-core A53 @ 2.3GHz
* Computing Power: Supports FP32/INT8; simultaneous decoding and analysis of 8 HD video streams
* Memory: 6 GB
* Storage: 32 GB
* Video Decoding: Supports 8x 1080P@30FPS streams
* Video Encoding: Supports 1x 1080P@25FPS stream
* Network Ports: 2 x 10/100/1000M adaptive Ethernet ports
* Power Supply: DC 12V
* Typical Power Consumption: 18W
* Interfaces: – 2 x USB 3.0 – 1 x HDMI – 1 x TF card slot – 2 x RS-485 – 1 x Digital I/O – 2 x Alarm Inputs – 1 x Audio – 1 x DC12V output
* Operating Temperature: -20°C to +70°C (varies by configuration)
