Skip to main content

UAV / Drone Computer Vision Systems

High-Performance Drone Computer Vision

Drone Computer for Imaging & Vision Applications

A drone computer must process data instantly. UAV missions demand low latency and high reliability. Therefore, Gidel FantoVision edge computers are built for real-time imaging at the edge. They combine a compact form factor, rugged construction, and FPGA-accelerated processing. As a result, UAVs can make fast decisions in the air.

High-Bandwidth Multi-Camera Support

Our Low SWaP systems are 13.4 × 9 × 6 cm and 750 g. They fit easily into multirotor, fixed wing, and VTOL platforms. They also support 10GigE Vision, CoaXPress, and Camera Link. This enables synchronized multi-camera capture up to 40 Gb/s throughput. Therefore, they are ideal for:

  • High-resolution imaging workflows
  • Aerial inspection and surveying
  • Real-time mapping and 3D reconstruction
  • Situational awareness, defense, and ISR
  • Search and rescue and environmental monitoring

FPGA + AI Hybrid Edge Processing in Drone Computers

FPGA pipelines handle deterministic real-time data flow while the NVIDIA Jetson performs onboard AI inference. This heterogeneous approach is designed to overcome common UAV vision bottlenecks such as excessive CPU load, unpredictable latency, bandwidth saturation between processing stages, and limited scalability as camera count or resolution increases. Up to 2 TB internal storage supports long autonomous missions and extended data capture when required.

Flexible Drone Computer Architecture and Real-Time Image Enhancement

All image signal processing runs inline on the FPGA to preserve deterministic, low-latency behavior. This includes debayering, non-uniformity correction (NUC), bad pixel correction (BPC), white balance, gamma correction, and dynamic luminance balancing. Gidel’s real-time HDR processing enhances visibility and contrast under challenging lighting conditions.

To optimize bandwidth and storage, the FPGA also supports real-time data reduction using JPEG, Lossless, and Quality+ compression, along with region-of-interest extraction and scaling. These capabilities reduce data movement and prevent downstream processing bottlenecks without compromising image quality.

Development and customization are supported through Gidel’s Software Development Tools, which enable developers to design, integrate, and optimize FPGA-based imaging and vision pipelines while maintaining full control over latency, determinism, and system behavior.

Gidel provides a scalable, low-latency drone computer for edge imaging. We also offer Mini Powerful FPGA modules for even tighter SWaP constraints or advanced customization. This gives UAV integrators full flexibility while ensuring mission success and real-time decision-making.

Need more information?




    Resources

    Join Reuven Weintraub, founder and CTO of Gidel, as he reveals how to unlock exceptional image processing performance by adding FPGA’s processing to NVIDIA Jetson edge AI systems.

    Reuven Weintraub – Gidel Founder and CTO describes how to boost performance in Machine Vision and Imaging systems. This talk has been given at the Vision Show 2024 Stuttgart, during the Vision Days Forum.

    Image processing acceleration on NVIDIA Jetson embedded computer at the Embedded World 2024 show.

    Gidel edge computers and frame grabbers at the Embedded World show 2024.

    A short video introduction to FantoVision edge computers – 2023.

    Need More Information?

    Our team of experts is ready to help you find the perfect solution for your application needs.

    • Free technical consultation
    • Custom solution design
    • Quick response time
    • Expert engineering support

    Or contact us directly:

    [email protected]




      Quote