Implementation Of Four Step Search Algorithm For Motion Estimation Using MATLAB

Elya Irwana , Zullkifly (2009) Implementation Of Four Step Search Algorithm For Motion Estimation Using MATLAB. Project Report. UTeM, Melaka,Malaysia. (Submitted)

	PDF (24 Pages) Implementation_Of_Four_Step_Search_Algorithm_For_Motoin_Estimation_Using_Matlab_Elya_Irwana_Bt_Zullkifly_QA76.9.A43.E49_2009_-_24_Pages.pdf - Submitted Version Download (641kB)
	PDF (Full Text) Implementation_Of_Four_Step_Search_Algorithm_For_Motoin_Estimation_Using_Matlab_Elya_Irwana_Bt_Zullkifly_QA76.9.A43.E49_2009.pdf - Submitted Version Restricted to Registered users only Download (7MB)

Abstract

This project presents an overview of Block Matching for Motion Estimation in video coding. The best solution of block matching implementation is Full Search (FS) Algorithm. Due to high computational workload required in this algorithm, various types of block matching algorithms have been proposed and developed to overcome this problem. The Four Step Search (4SS) Algorithm has been proposed for the study in this project. This algorithm features are based on the centre-biased motion vector distribution characteristic and contents halfway-stop techniques to reduce computation process. This proposed algorithm contains 1 to 4 searching step and 17 to 27 checking point. From the simulation result using Matlab, most of performance in term of PSNR of this algorithm better than others; Cross Search (CS) Algorithm, New Three Step Search (NTSS) Algorithm, Diamond Search (DS) Algorithm and Cross Diamond Search (CDS) Algorithm, and has close to FS Algorithm.

Actions (login required)

View Item