GPU Programming
As part of real-time data processing, Shenandoah Solutions can design and implement algorithms that utilize Graphics Processing Units (GPUs).
GPUs provide a highly parallel processing environment and we develop software that takes advantage of that environment to process voluminous or high-rate data.
Typical domains applicable for GPUs include:
- Digital Signal Processing (DSP)
- Image Processing
- Modeling Applications
Implementation on GPUs is a software approach that is often less expensive, faster to implement, and more easily maintainable than a Field Programmable Gate Array (FPGA) firmware implementation.
Some of examples of GPU image processing
Demosaicing
![](/images/boat.jpg)
![](/images/boat_demosaic.jpg)
Non-Uniformity Correction, Color Balance, and Stitching
![](/images/dock.jpg)
![](/images/dock2.jpg)
Image Stabilization
![](/images/stabilize.jpg)
![](/images/stabilized.jpg)
Panorama and Close Up
![](/images/panorama.jpg)