Design and implementation of an embedded H.264 color video encoding pipeline for a mobile processing platform

Date of Award

2016

Degree Name

M.S. in Electrical Engineering

Department

Department of Electrical and Computer Engineering

Advisor/Chair

Advisor: Eric John Balster

Abstract

In this thesis, development of a real time H.264 embedded color video encoding pipeline for a mobile processing platform is proposed. Also, a rate-distortion analysis of the optimum color video compression mode using H.264 is determined. The mobile processing pipeline addresses the growing need of viewing high definition video on mobile devices, such as a tablet or phone in a low bandwidth environment. The advancements in today's hardware technology allow this problem to be solved through a compilation of open source software systems and low size, weight, and power (SWAP) hardware. There are over six color video formats to use within H.264, so a rate-distortion analysis is required to determine the optimum color video format. The results show that for 1080p video, the YUV 4:2:0 video with a chrominance quantization parameter offset of zero is among the best performing color video format in terms of rate-distortion quality for three different levels of compression and two different entropy encodings. Using that result an end-to-end color video processing pipeline is created around an embedded low power H.264 encoder to provide 720p color video to an end user. Using this encoding pipeline, there is an observed 2x improvement of rate-distortion quality over the currently implemented system based upon JPEG compression, as well as a 2x timing improvement for 720p video. Throughput testing highlights the capabilities of the processing pipeline, and the end goal of broadcasting a 720p color video stream is satisfied.

Keywords

High definition video recording, Data compression (Telecommunication), Rate distortion theory, Signal processing, Cell phones, Tablet computers, Streaming video, Electrical Engineering, H264-Compression-GST-YUV

Rights Statement

Copyright © 2016, author

Link to Full Text

Share

COinS