FPGA Image Signal Processor

Key Features:

• Real-Time Image Processing in Hardware
  Offloads critical ISP tasks to FPGA, minimizing CPU/GPU usage and improving performance.

• Modular Architecture
  Mix and match the following hardware-accelerated modules:

  • Debayering

  • Color Space Conversion (BGRA ↔ UVVY)

  • Noise Filtering

  • Auto White Balance

  • Histogram Equalization

• Standard V4L2 Interface
  Each module is presented as a V4L2 device—no custom interface or user-space adaptation required.

• Built for Embedded Vision Pipelines
  Designed to operate as standalone accelerators or part of a complete GStreamer pipeline.

Example Use Case:

An industrial inspection system using a Xilinx UltraScale+ FPGA and NVIDIA Jetson Xavier NX captures raw Bayer-format video from a high-resolution camera. Using RidgeRun’s FPGA ISP, the system performs real-time debayering, noise filtering, and color conversion in hardware—offloading the CPU and delivering clean, preprocessed frames to an AI inference engine running on the Jetson. This allows the system to process video at full frame rate with minimal latency.

What’s Included:

• Pre-synthesized FPGA ISP modules

• Support for integration with RidgeRun’s V4L2 FPGA framework

• Example pipelines and developer documentation

• Optional 11-hour Embedded Software Support package

Supported Platforms:

• NVIDIA Jetson (Nano, TX2, Xavier NX, AGX Orin)

• NXP i.MX6 / i.MX8

• Xilinx UltraScale+, Zynq MPSoC

• Custom FPGA boards supported—contact us for support

Why RidgeRun?

With over 19 years of experience in embedded multimedia systems, RidgeRun delivers production-grade solutions that balance real-time performance, low power consumption, and seamless integration. The FPGA ISP lets you build scalable, reliable video processing pipelines ready for deployment in vision systems, robotics, automotive, and industrial AI.

Learn more in our developer wiki