Title	Matlab programming for 2D autocorrelation analysis
Authors	Tsang, Wai Lek
Advisors	Yu, Siu Fung
Degree	Undergraduate
Issue Date	2018
Absctract	Autocorrelation is a mathematical operation performed on two functions to find a third one. It can be described as the cross-correlation of a signal with itself but adding some amount of time shift. The technique of Digital image correlation (DIC) is commonly used in digital signal and image analyses because it is useful for finding out a certain pattern in one signal. The aim of this project is to study and attempt the application of autocorrelation in 2D image analysis through programming in Matlab, mainly focus on template matching and object tracking. Two types of autocorrelation were investigated in this project; which were done in spatial domain and frequency domain respectively. The results showed that while autocorrelation is capable for object tracking and movement measurement in image or video format, its accuracy largely depends on the image quality of tracking target during each frame. In order to acquire a highly accurate tracking result, it is essential to have high image quality and precise pixel correspondence.
Subjects	MATLAB Image processing Autocorrelation (Statistics)
Publisher	Hong Kong Polytechnic University
Rights	All rights reserved.