Real-Time RAW to JPG for Aerial Imaging
- Available for all Gidel’s FPGA accelerator hardware platforms
- Supports High-End GigE Vision, CoaXPress and Camera Link cameras
- Processes over 7 images per second for massive sensor data
- Supports HRS up to ~500MP per image / 32K width
- Real-Time ISP for Enhanced Image Quality including:
- High-Quality Demosaicing
- High Dynamic Range (HDR IP) correction
- Chromatic Aberration, NUC, and BPR
- RAW (Bayer) to JPG compression exceeding 10:1
- Support camera control via GenICam GenTL and Integrated with the Gidel Grabber SDK suite
- Optional: Full ISP pipeline customization using the ProcVision SDK
- H.264 / H.265 & RTSP output Options
Product Overview
SkyBoost-RT is a dedicated solution for real-time aerial imaging and similar applications, designed to accelerate End-to-End processing directly on mission-critical airborne platforms. It enables High-Quality on-the-fly Image Signal Processing (ISP) and compression while the HRS is actively capturing RAW data, allowing high-quality imagery to be processed, recorded, or streamed in real time for Military & Defense ISR, Search & Rescue, and Situational Awareness.
Built for pipelines where deterministic behavior, low latency, and continuous operation are critical, the solution processes and compresses RAW image data immediately during acquisition. This ensures seamless data flow without buffering delays or the need for intensive post-processing.
Solving Data Bottlenecks in Aerial Imaging
In many airborne systems, storing or transferring uncompressed RAW image data is not practical. High-resolution sensors (HRS) quickly generate terabytes of data, creating significant challenges for storage capacity, bandwidth, and real-time usability.
SkyBoost-RT addresses these challenges by performing image processing and compression on the fly:
- The system processes High-Resolution RAW (Bayer) image data in real time while the camera is filming
- FPGA-based ISP enhances image quality before compression
- JPG compression significantly reduces data volume during acquisition, often exceeding a 10:1 compression ratio
- Deterministic behavior ensures continuous operation without dropped frames
Furthermore, this approach enables long-duration recording without accumulating massive storage requirements and makes real-time streaming substantially easier due to the reduced data bandwidth.
High-Quality Processing Pipeline Optimized for Aerial Imaging
An FPGA-based image processing pipeline optimized for aerial imaging and real-time operation performs all stages deterministically. Specifically, the pipeline includes:
- Demosaicing – Converts raw Bayer data into sharp, full-color RGB imagery
- High Dynamic Range (HDR) – Captures superior details in high-contrast lighting conditions
- Chromatic Aberration Correction – Corrects color fringing and lens-related color misalignments for crisper edges
- Non-Uniformity Correction (NUC) – Calibrates individual pixel responses to ensure a uniform image across the entire sensor
- Bad Pixel Replacement (BPR) – Automatically identifies and compensates for defective pixels in real time
- Dynamic Luminance Balance – Preserves consistent brightness under changing illumination
- White Balance – Maintains color accuracy across variable lighting conditions
- JPG compression exceeding a 10:1 ratio directly on the FPGA
- Optional: H.264 / H.265 – Industry-standard video compression for efficient recording and transmission
- RTSP Output Options – Enables bandwidth-efficient video streaming and recording
All processing stages operate with controlled buffering and predictable latency, ensuring stable performance under sustained high data rates.
Throughput for High-Resolution Aerial Sensors (HRS)
The architecture is purpose-built for Ultra-Aerial-HRS, supporting image widths of up to 32K pixels. This corresponds to cameras of up to ~500MP, depending on the model and configuration.
In real-time pipelines, the system can process over 7 images per second, depending on configuration and input format. Additionally, it sustains high frame rates even at very large frame sizes, enabling continuous acquisition, processing, and compression without interrupting the image stream.
Performance at Scale
| Sensor Class | Processing Speed | Throughput |
|---|---|---|
| 100 MP | > 7 FPS | > 2 Terapixels / Hour |
| 150 MP | > 4.5 FPS | > 2 Terapixels / Hour |
*Throughput values depend on specific pixel format and system configuration.
Deployment on Airborne and Embedded Platforms
Gidel supports deployment across multiple hardware platforms, depending on system architecture and operational constraints:
- Mini Jetson Frame Grabber Systems combining FPGA acceleration with embedded GPU processing
- FPGA Frame Grabbers installed in user PCs or servers
- FPGA modules for compact, embedded, and low-SWaP deployments
This makes the solution particularly well suited for platforms where size, weight, and power matter, such as embedded airborne systems, compact payloads, and the broader aerial imaging environment.
Supported Interfaces
The system supports a wide range of high-end camera and sensor interfaces, including:
- 1-10 GigE Vision
- CoaXPress-12 and CoaXPress-6
- All Camera Link configurations
- Custom user-defined protocols
ISP Customization for UAV and Aerial Systems
The ProcVision SDK provides full control over the Image Signal Processing pipeline through FPGA-level customization. This enables application-specific ISP design to meet precise imaging requirements.
Additionally, the system calibrator handles pipeline optimization, allowing configuration of the processing flow to balance image quality, compression ratio, throughput, and system resource utilization.
Why Choose SkyBoost-RT for Your Aerial Imaging Application?
SkyBoost-RT overcomes the storage and bandwidth limitations of airborne platforms by processing and compressing massive RAW data streams in real time, ensuring you capture all images without overwhelming your downlink. It empowers mission-critical ISR and inspection operations with immediate visual intelligence, all within the tight power and weight limits of modern UAVs.
Need to accelerate your workflow back at the office instead?
Discover SkyBoost to drastically reduce your post-processing times.
General
- 1-10 GigE Vision
- CoaXPress-12 / CoaXPress-6
- Camera Link
- Custom User Protocols
- Bayer, Mono (8, 10, 12, 14 and 16 bits/color)
- Altera Stratix 10 (NX/MX/SX)
- Arria 10 GX
Performance
| Frame Size | FPS |
|---|---|
| 100MP | > 7 |
| 150MP | > 4.5 |
| Metric | Capacity per Hour |
|---|---|
| Pixel Count | > 2 Terapixels |
The SkyBoost-RT is highly modular!
Want to adjust it according to your Vision?
| Name | Description | Type | Size | |
|---|---|---|---|---|
|
SkyBoost-RT
|
Datasheet | 1.8 MB | ||
|
SkyBoost-RT
|
Block Diagram | JPG | 119.5 KB | |
|
SkyBoost-RT
|
Discover how to overcome critical High-Resolution Outdoor Imaging challenges using Gidel’s ISP and transform massive... | Link | — | Read |
FAQ
SkyBoost-RT addresses the critical data bottlenecks of airborne platforms where high-resolution sensors generate terabytes of RAW data. By performing on-the-fly Image Signal Processing (ISP) and compression, it reduces data volume by over 10:1 during acquisition. This enables real-time recording and streaming of high-quality imagery without overwhelming storage capacity or downlink bandwidth.
Gidel supports deployment across a variety of hardware architectures to meet different operational constraints:
-
Mini Jetson Frame Grabber Systems: Combining FPGA acceleration with embedded GPU processing (ideal for Edge AI).
-
PCIe FPGA Frame Grabbers: Installed in standard user PCs or airborne servers via PCIe.
-
FPGA Modules: Such as the FDB series, for ultra-compact, low-SWaP (Size, Weight, and Power) deployments in small UAVs and gimbals.
The architecture is purpose-built for Ultra-High-Resolution (HRS) sensors, supporting image widths up to 32K pixels and sensors up to ~500MP. It sustains a massive processing volume of over 2 Terapixels per hour, achieving speeds of >7 FPS at 100MP and >4.5 FPS at 150MP, ensuring continuous acquisition without dropping frames.
Unlike standard compression tools, SkyBoost-RT utilizes a sophisticated FPGA-based ISP pipeline that includes Demosaicing, Chromatic Aberration Correction, and Bad Pixel Replacement before the compression stage. This “pre-cleaning” of the RAW data ensures that the resulting JPEG or H.264/5 stream maintains professional-grade clarity and detail, even at high compression ratios.
Absolutely. We provide the ability to add image algorithms according to specific customer requirements. In addition, the user can utilize the system calibrator to manage pipeline optimization, allowing for precise control over the processing flow to balance image quality, compression ratio, and throughput based on the unique needs of the mission.
SkyBoost-RT offers extensive flexibility, supporting high-end camera interfaces including 1-10 GigE Vision, CoaXPress-12, and all Camera Link configurations. Additionally, the system can be configured to support custom user-defined protocols, allowing for seamless integration with specialized or proprietary sensors.
SkyBoost-RT is designed for immediate visual intelligence. It offers RTSP output options, enabling bandwidth-efficient video streaming and recording. This allows ground teams to view high-quality, processed video feeds in real-time while the mission is still in the air, a critical feature for ISR and Search & Rescue operations.