News & Events
Cincoze GM-1000 Performs Computer Vision for Thermal Imaging Systems
Introduction
COVID-19 is not over yet. So, to decrease its spread, many public health authorities worldwide have turned to machine vision and imaging technologies to build front line infection tracking systems. One of our customers in China set up thermal imaging cameras for fast individual screenings in airports, hospitals, factories, office buildings, restaurants, and stores to prevent virus transmission.
Our customer wanted to develop automated thermal imaging systems. These systems provide real-time body temperature indicators through contactless thermal monitoring and biometric data analysis, allowing on-site personnel to quickly identify individuals with potential illnesses. Combined with AI and deep learning algorithms, it can also perform mask detection, face recognition, image search, footprint tracking, and other functions, providing real-time information for the intelligent analysis of big data in the central control center. An industrial GPU computer is essential because of its GPU computing performance for inference analysis, enabling large-scale screening to eliminate coronavirus outbreak and other infectious diseases.
Customer's Requirements
GPU Accelerated Computing
Thermal cameras capture infrared images of passengers, then send that data to GPU computers. The dedicated GPU must execute the intensive computer vision algorithms while reserving CPU power for system management, control, and communication tasks. Fast image processing and analysis can avoid the bottlenecks that result in long wait queues and crowds in public venues.
Small yet Versatile
The computer needs to be compact to fit in a small customized enclosure. Rich I/O connectivity to sensors, cameras, other devices, and networking are required. It also needs to be scalable to expand its solution to address some other circumstances.
Built for Harsh Environments
The thermal detection system is an edge deployment. Therefore, an industrial-grade GPU computer is required. The computer supports -40–70°C wide operating temperature ranges and has a robust aluminum case. Effective heat dissipation design is also vital to ensure GPU computing with reliability.
Why Cincoze?
Embedded MXM GPU Computer
The Cincoze GM-1000 supports MXM 3.1 socket Type A/B expansion modules, including MXM GPU modules based on AMD Radeon™ E9174, NVIDIA® Quadro® P2000, T1000, and RTX3000 GPUs. Customers can freely upgrade the MXM modules to suit different application processing requirements in machine vision, image processing, and artificial intelligence.
Compact and Rich Functions
The GM-1000 is compact and is highly integrated, measuring only 260 x 200 x 85 mm. It provides general industrial I/O including 4 x COM, 2 x GbE LAN, 8 x USB, 1 x HDMI, and 1 x DVI-I and is available with additional customizable expansion options. Through Cincoze’s exclusive CMI and CFM technologies, GM-1000 can accommodate more interfaces and functions such as 2 x 10GbE LAN, 4 x GbE LAN, PoE, and ignition sensing.
Ruggedness
The GM-1000 packs performance and flexibility in a robust and durable package. In addition to the fanless design (optional external fan kit available), the GM-1000 is crafted with industrial-grade components, prominent aluminum fins, heat sinks, and heat spreaders. It passed a series of rigorous tests such as wide operating temperature (-40–70°C), vibration (5G), and shock (50G) tests. It is also certified to EN62368-1 safety and EN 50121-3-2 rolling stock standards.
