HawkEye-CXP12:

Quad CoaXPress-12 Frame Grabber

4 x CXP-12 Frame Grabber & Image Processing System

Up to 50 Gb/s bandwidth via 4 x CoaXPress-12 links
Ideal for Recording & Streaming solutions
Inline FPGA image processing options include:
- Real-time detection for object, event, or feature identification
- On-FPGA real-time compression: JPEG, Lossless, Quality+
- Real-time High Dynamic Range (HDR) correction
Supports 100+ synchronized cameras using the InfiniVision
Memory: up to 17 GB
Form factor: full-height, single-slot, half-length PCI Express card
Optional: Low Profile

Quad CoaXPress-12 Frame Grabber

The HawkEye-CXP12 is a high-performance Quad CoaXPress-12 Frame Grabber designed for real-time image acquisition, preprocessing, and compression in demanding multi-camera vision systems. Powered by Altera Arria 10 FPGA technology, it delivers up to 50 Gb/s bandwidth via 4 × CXP-12 links. As a result, it ensures zero frame loss, ultra-low latency, and negligible CPU load, making it ideal for mission-critical applications.

Quad CXP-12 Acquisition with Real-Time FPGA Image Processing

The HawkEye-CXP12 is available as a plug-and-play CoaXPress frame grabber or as part of a complete Imaging & Vision system with real-time FPGA preprocessing, image enhancement, and compression options.

The built-in FPGA can process image data during acquisition, before it reaches the host PC. This helps reduce bandwidth, storage, and host processing load while preserving low-latency performance for high-speed camera streams.

Optional FPGA processing includes Compression, HDR correction, Detection, and additional image enhancement modules.

Real-Time FPGA Processing Helps Enable:

Reduced bandwidth for high-throughput image acquisition
Low-latency processing directly in the acquisition flow
Extended recording time through real-time compression
Lower host processing load by offloading selected tasks to the FPGA
Custom image pipelines using Gidel IPs or user FPGA logic

For the full list of available image processing and enhancement options, please refer to the Options tab.

Flexible Operating Modes

The HawkEye-CXP12 can switch between InfiniVision Mode and ProcFG Mode via firmware update:

InfiniVision: Ideal for synchronized multi-camera setups, combining all camera data—even across multiple cards—into a single buffer with support for dynamic resolutions and formats.
ProcFG: Optimized for line-scan cameras, offering fixed frame sizes, pixel formats, and ROI grabbing (uncompressed).

As a result, the HawkEye-CXP12 adapts seamlessly to various workflows, supporting everything from dynamic multi-camera systems that require precise synchronization to streamlined single-camera operations demanding consistent performance.

InfiniVision: Multi Camera Acquisition and Synchronization

Gidel’s InfiniVision architecture addresses key multi-camera challenges, including bandwidth, synchronization, connectivity, and scalability. As a result, the HawkEye-CXP12, Quad CoaXPress-12 Frame Grabber supports up to four CXP-12 links, enabling 1–4 cameras to operate simultaneously with precision.

In addition, a PCIe Gen.3 x8 interface delivers CPU-free, ultra-fast data transfer, while on-board buffers of up to 16 GB boost acquisition stability and real-time image processing. Consequently, users achieve smooth, reliable performance in high-speed, high-resolution systems.

SDK and Development Tools

The HawkEye-CXP12 is supported by Gidel’s SDK, which includes intuitive GUIs and APIs for streamlined integration. In addition, the ProcVision Suite provides FPGA programming, debugging, and validation tools, enabling rapid customization of data flows, image processing, and compression pipelines. Consequently, developers can create optimized, application-specific solutions faster and with reduced risk.

Why Choose the HawkEye-CXP12 Frame Grabber?

Quad CXP-12 FPGA frame grabber for high-speed multi-camera image acquisition
Up to 50 Gb/s aggregate input bandwidth
FPGA architecture for low-latency acquisition, processing, recording, and streaming
Flexible operating modes for multi-camera and precision setups
Advanced SDK and ProcVision Suite for rapid development and integration
Optional real-time FPGA processing, image enhancement, and compression IPs

The HawkEye-CXP12 is the ideal choice when your workload requires high-bandwidth CXP-12 acquisition, low-latency FPGA processing, and reliable image transfer to a host PC.

General

Supported Protocol

CoaXPress 2.1

Camera Input Options

4 x CoaXPress-12 links
2 x CoaXPress-12 links
1 x CoaXPress-12 link
PoCXP

Camera Input Connectors

4 x Micro-BNC

Supported CXP Downlink Speeds

1.25 Gb/s | CXP-1
2.5 Gb/s | CXP-2
3.125 Gb/s | CXP-3
5 Gb/s | CXP-5
6.25 Gb/s | CXP-6
10 Gb/s | CXP-10
12.5 Gb/s | CXP-12

Supported CXP Uplink Speeds

20.83 Mbps (CXP-1 to CXP-6)
41.66 Mbps (CXP-10, CXP-12)

Maximum Stream Packet Size

8,192 bytes

Additional Connectors

JTAG
VGA15-pin for connecting GPIO's to the bracket

Advanced ROI feature

On-the -fly selective ROI acquisition

Image Formats

Mono, Bayer, RGBA (8, 10, 12, 14 and 16 bits/color)
RGB (8, 10 and 12 bits/color)

Max Resolution

Horizontal: 16 K pixels (64-bit)
Vertical: 65 K lines

Maximum Acquisition Throughput

40 Gb/s

Real-Time FPGA Image Compression

Supports optional FPGA-based compression options.
For the full list, please refer to the Options tab.

Compression & Image Enhancements

Compression Options:

Image Enhancements Options:

High Dynamic Range (HDR) correction

For more modular image-processing options - Contact Us

Real-Time FPGA Image Processing & Enhancements

Supports optional FPGA image processing and enhancement modules.
For the full list, please refer to the Options tab.

Host Bus

PCIe x8 Gen. 3

On Board Memory

Up to 17 GB

Camera Types

Area
Line Scan

Form Factor

Full-height, single-slot, half-length PCIe card

GPIO

GPIO #1:

4 x RS422 INPUTS
Optional input clk
1 x Optocoupler input

GPIO #2:

1 x RS422 INPUTS
2 x Opto-coupler inputs
2 x 3-30V @ 0.8A outputs
LVTTL IO
12V/1A power supply

GPIO #3: (for board-to-board connection):

12 x LVTTL IO (5V TTL tolerant)

Power

5-32W: Depends on user application and FPGA Type

Cooling

Passive cooling
Active cooling (Fan)

MTBF

Passive colling: 1.7M hours
Active colling: 350K

Environmental conditions

Temperature

Operating ambient air temperature: 0 – 55° C

Humidity

Continuous Operation: 10 - 80% (non-condensing)
Peak Operation: 10 - 90% (non-condensing)

Environmental Compliance

Real-Time FPGA Image Compression

Real-Time FPGA Image Processing & Enhancements

HDR Correction – High Dynamic Range correction from a single exposure
Detection – Enables real-time object, event, or feature detection at the acquisition stage
Chromatic Aberration Correction – Reduces color fringing caused by optical misalignment
Dynamic Luminance Balance – Preserves consistent brightness under changing illumination
White Balance – Maintains color accuracy across variable lighting conditions
NUC – Non-uniformity correction for improving sensor response consistency
BPR – Bad pixel replacement for correcting defective or unstable pixels
Morphological operations – Open/Close operations using a round structuring element

User FPGA code

Users can integrate their own FPGA code alongside Gidel IPs.

For more information, refer to the FPGA Processing tab

Power

PoCXP 32W
PoCXP 42W
PoCXP 52W

CXP Links

1 x CXP-12 link
2 x CXP-12 links
4 x CXP-12 links

Form Factor

Low profile (Comes without the IO connector)

The HawkEye-CXP12 Frame Grabber is a highly modular solution, designed to be tailored to meet unique application requirements.

Looking to adapt the HawkEye-CXP12 to match your vision?

Contact Our Experts

The HawkEye-CXP12 frame grabber offers two powerful customization paths to meet specific application requirements:

Pre-Configured FPGA Features
Gidel develops a range of pre-embedded modules tailored to your specifications, including real-time compression, image processing, and enhancement options.
For the full list of image processing and enhancements, please refer to the Options tab.
User-Level FPGA Customization
Leverage Gidel’s development tools and IP libraries to integrate your proprietary FPGA logic and extend the system’s capabilities for acquisition, image processing, and control.

The Gidel ProcVision Suite delivers a complete toolchain for advanced user-level FPGA customization of the data flow, image pipeline, image processing, and more, ensuring optimal performance for your vision or imaging application.

Gidel can also integrate customer algorithms into the FPGA or develop custom FPGA logic according to specific application requirements.

HawkEye-CXP12: FPGA resource comparison

Model	HawkEye-CXP-48	HawkEye-CXP-16
FPGA	Arria 10 480 GX	Arria 10 160 GX
FPGA-ALM	182K	61K
FPGA-M20K	1,431	440
FPGA - 18 x 19 Multipliers	2,736	312
DRAM A capacity	1 GB	1 GB
DRAM A sustain bandwidth	5.6 GB/s	6.4 GB/s
SoDIMM B capacity	0,4,8,16 GB	-
SoDIMM B sustain bandwidth	9.6 GB/s	-

Grabbers SDK

InfiniVision

Designed for acquisition from a large number of cameras (100+), with an option for embedded real-time compression.

ProcFG

Optimized for line-scan camera acquisition, combining ROI-based grabbing with integrated debugging and analysis tools.

Application Interfaces

GUI Applications

InfiniVision
ProcFG
CameraConfig – Camera discovery and configuration
ggvcon – GigE Vision network configuration

APIs

InfiniVision with supporting examples
ProcFG with supporting examples
Gen<i>Cam GenTL producer libraries compatible with C/C++ compilers
InitCam for developing user Gen<i>Cam camera configuration application
GigE for developing camera network communication applications

Software Compatibility

Third-party software

MVTec Halcon machine vision software
Camera control Gen<i>Cam based application

Operation Systems supported

Windows 11
Windows 10
Windows Server 2022
Windows Server 2019
Windows Server 2016
Linux (kernel 2.6.x- 6.12)

Please note: Linux version doesn’t include the ProcFG/InfiniVision GUI, just the API.

Name	Description	Type	Size
HawkEye-CXP12	Datasheet	PDF	278.6 KB
HawkEye-CXP12	Block Diagram	PNG	63.8 KB

Reuven Weintraub, founder and CTO of Gidel, reveals how to unlock exceptional image processing performance by adding FPGA’s processing at Vision Show, Stuttgart 2024.

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

Boost your vision! Gidel's high-performance GigE Vision, Camera Link and CoaXPress frame grabbers help overcome bandwidth bottlenecks in high-resolution and/or high-speed computer vision applications.

Gidel presented its FPGA-based architecture capable of processing Giga+ Pixels per second while maintaining exceptionally low power consumption. The session demonstrated how Gidel’s scalable FPGA solutions deliver real-time imaging performance, energy efficiency, and deterministic throughput for advanced vision and imaging systems.

Presented by Reuven Weintraub, this talk highlighted Gidel’s expertise in real-time processing over Gigapixel/s image streams, demonstrating how FPGA-based architectures enable deterministic latency, scalable throughput, and efficient handling of ultra–high-resolution vision data.

Gidel founder and CTO Reuven Weintraub present the performance and versatility of Gidel's GigE Vision, Camera Link and CoaXPress frame grabbers for high-speed image acquisition and pre-processing at the Embedded World 2022 in Nuremberg, Germany.

Gidel demonstrated a complete real-time object tracking pipeline implemented directly on the FPGA-based frame grabber, eliminating host dependency and overcoming PCIe bandwidth limitations. The demo showcased deterministic low-latency processing, highlighting Gidel’s ability to execute full vision workflows on FPGA hardware.

Gidel presented a step-by-step demonstration of its FPGA-based object tracking algorithm, highlighting the internal processing stages and real-time debugging capabilities. The demo illustrated how developers can visualize, analyze, and optimize the complete tracking flow directly on FPGA hardware, enabling faster development and deterministic performance.