Enhanced Image Processing beyond Baseband: CPU/GPU Process Model Unlocks Performance Possibilities

Ernie Sanchez, Kirk Marple, Hank Frecon

The speed and quality of traditional baseband approaches to image enhancement is now being eclipsed by commodity enterprise computing technology, with emerging file-based solutions focused on frame-rate conversions, de-noising, resolution-scaling, and other techniques for improving image quality. This paper will examine how continually increasing CPU power combined with increasingly sophisticated GPU-based algorithms enables ever-greater performance and output quality, with the significant operational and cost benefits found only in highly automated and massively parallelized grid-based processing workflows.

Published
2013-10
Content type
Original Research
Keywords
file-based, image-enhancement, computing technology, IT, baseband, CPU, GPU, processing, parallel, grid-based, frame-rate conversion, standards conversion, image processing, algorithms, transcoding
DOI
10.5594/M001529
ISBN
978-1-61482-953-9