VISION 2026 (Oct 6-8)
FPGA-Based Image Processing Framework for High-End Multi-Camera Applications
Engineered for low-latency. Designed to scale. Your framework for high-end multi-camera applications
Bonn/Germany, August 2026 - N.A.T. will be exhibiting at VISION 2026, the world’s leading trade fair for machine vision. The event takesplace from October 6–8, 2026, at Messe Stuttgart, Germany.
This year at booth 10H46, N.A.T. is putting the spotlight on NATvision, its framework for FPGA-based image processing, designed to provide a powerful foundation for demanding, high-end multi-camera applications.
"Engineered for low-latency. Designed to scale. Your framework for high-end multi-camera applications”
Under this slogan, N.A.T. will showcase the technology behind NATvision. The FPGA-based framework provides deterministic performance and data throughput required for modern vision systems – while maintaining the flexibility needed to adapt to individual applications.
NATvision: a framework for FPGA-based vision processing
At the heart of NATvision are N.A.T.’s in-house GigE Vision and CoaXPress (CPX) stacks, providing the infrastructure for high-speed image acquisition directly in the FPGA. Combined with an integrated GenICam interface, NATvision provides the building blocks for connecting GigE Vision and CoaXPress cameras and implementing application-specific image-processing pipelines on FPGA-based platforms.
NATvision allows developers to take advantage of FPGA technology while retaining the flexibility to create customized vision architectures. It allows to implement image acquisition, preprocessing and application-specific processing close to the point of acquisition, helping to minimize data movement and processing latency. When designing the framework, special care has been taken to allow for an easy exchange of any of the GigE Vision and CoaXPress stacks against other stacks.
Engineered for low latency
High-end multi-camera applications generate enormous amounts of image data. They also require fast and predictable processing. NATvision addresses these requirements by bringing image acquisition and processing together within the FPGA.
Processing image data directly in programmable logic enables deterministic, low-latency data paths. It reduces the need to transfer large volumes of raw image data between separate processing components.
This approach is particularly suitable for applications where predictable response times and high-throughput data handling are essential and where conventional architectures can quickly encounter limitations in data movement, processing latency and scalability.
Designed to scale
Modern vision systems are continuously evolving toward higher resolutions, faster frame rates and increasing numbers of cameras. NATvision is designed to provide a scalable framework that can evolve with these requirements.
Its modular architecture allows FPGA resources and interfaces to be adapted to the specific needs of an application. FPGA-based image preprocessing can also be combined with CPU, GPU or AI acceleration technologies, allowing processing tasks to be distributed according to the overall system architecture.
This makes NATvision a suitable foundation for demanding applications such as industrial inspection, quality control, high-speed measurement, scientific imaging, testing and other data-intensive multi-camera systems.
From GigE Vision and CoaXPress to application-specific processing
With its GigE Vision and CoaXPress stacks with GenICam interface, NATvision provides developers with essential camera-interface functionality as part of the framework. This substantially reduces the effort required to build the underlying infrastructure for an FPGA-based vision system.
On this foundation, developers can implement application-specific functions such as image preprocessing, filtering, feature extraction or other algorithms directly in the FPGA. The resulting data can then be passed efficiently to subsequent processing stages, including CPU, GPU or AI-based processing.
NATvision therefore combines high performance, deterministic behavior and application-specific flexibility in a framework designed for high-end multi-camera vision applications.
Meet N.A.T. at VISION 2026, booth 10H46
Visitors to VISION 2026 are invited to meet the N.A.T. team and discuss their requirements for low-latency, high-throughput and scalable multi-camera image-processing systems.
The focus at this year’s exhibition will be on NATvision and the technology framework behind the applications. Joing us at our booth 10H46 provides an opportunity for in-depth discussions about FPGA-based architectures, GigE Vision, GenICam integration and customized image-processing solutions.
We are looking forward to seeing you on-site at VISION 2026, the world’s leading trade fair for machine vision
N.A.T.
Engineered for low-latency. Designed to scale. Your framework for high-end multi-camera applications.