Award Date

1-1-2008

Degree Type

Thesis

Degree Name

Master of Electrical Engineering (MEE)

First Committee Member

Henry Selvaraj

Number of Pages

58

Abstract

Motion Estimation in MPEG (Motion Pictures Experts Group) video is a temporal prediction technique. The basic principle of motion estimation is that in most cases, consecutive video frames will be similar except for changes induced by objects moving within the frames. Motion Estimation performs a comprehensive 2-dimensional spatial search for each luminance macroblock (16x16 pixel block). MPEG does not define how this search should be performed. This is a detail that the system designer can choose to implement in one of many possible ways. It is well known that a full, exhaustive search over a wide 2-dimensional area yields the best matching results in most cases, but this performance comes at an extreme computational cost to the encoder. Some lower cost encoders might choose to limit the pixel search range, or use other techniques usually at some cost to the video quality which gives rise to a trade-off; Such algorithms used in image processing are generally computationally expensive. FPGAs are capable of running graphics algorithms at the speed comparable to dedicated graphics chips. At the same time they are configurable through high-level programming languages, e.g. Verilog, VHDL. The work presented entirely focuses upon a Hardware Accelerator capable of performing Motion Estimation, based upon Block Matching Algorithm. The SAD based Full Search Motion Estimation coded using Verilog HDL, relies upon a 32x32 pixel search area to find the best match for single 16x16 macroblock; Keywords. Motion Estimation, MPEG, macroblock, FPGA, SAD, Verilog, VHDL.

Keywords

Accelerator; Based; Bma; Estimation; Hardware Implementation; Motion

Controlled Subject

Electrical engineering

File Format

pdf

File Size

1.87 MB

Degree Grantor

University of Nevada, Las Vegas

Language

English

Permissions

If you are the rightful copyright holder of this dissertation or thesis and wish to have the full text removed from Digital Scholarship@UNLV, please submit a request to digitalscholarship@unlv.edu and include clear identification of the work, preferably with URL.


Share

COinS