Optical flow estimation based on slantlet transform

Dissertant

al-Ruby, Hamsa Diya Majid

Thesis advisor

Sabri, Athir A.

University

University of Technology

Faculty

-

Department

Department of Electrical Engineering

University Country

Iraq

Degree

Master

Degree Date

2012

English Abstract

A model called Slantlet Based Optical Flow Estimation (SLT_OFE) is proposed in this thesis to estimate the optical flow.

Slantlet Transform (SLT) used as an effective tool, 2D and 3D- SLT- Level 2 are computed and employed in the proposed model.

Through this work, it is found that using SLT provides more accurate estimation of the optical flow.

Furthermore, SLT reduces the computationally complexity by giving a sample reduction in its filterbank by (2L -2) for each level L, and compact the signal energy into a small number of coefficients leading to reduce the program implementation time and the size of required memory.

By its definition, optical flow is a velocity field.

So, the velocities in this thesis are computed using the widely used Differential Technique, two methods from this technique are adopted, Horn-Schunck Method and Lucas-Kanade Method.

For extreme study of its performance, the proposed SLT_OFE is compared with three other models that are based on Discrete Wavelet (DWT2), Discrete Multi-Wavelet (DMWT2) and Framelet (FT2) level two transformations which are implemented in this work and employed in the conventional models, 2D-DWT2_OFE, 2D-DMWT2_OFE and 2D-FT2_OFE.

All of these models are tested under the same conditions by using two types of image sequences, synthetic and real image sequences.

Unlike the real sequences, the synthetic sequences have known true velocities which are used for the evaluation.

The optical flow is assessed by calculating mean error (Mean Err.), angular error (Ang.

Err.), and Standard Deviation (STD).

The results show that SLT_OFE gives minimum values in errors and STD using 2D-SLT2 and these results are improved by using 3D-SLT2.

This leads to the fact that the proposed model SLT_OFE through both of 2D-SLT2_OFE and 3D-SLT2_OFE approaches gives an improvement in the optical flow estimation process with high accuracy than the other models produced in the same circumstances.

Main Subjects

Electronic engineering

Topics

American Psychological Association (APA)

al-Ruby, Hamsa Diya Majid. (2012). Optical flow estimation based on slantlet transform. (Master's theses Theses and Dissertations Master). University of Technology, Iraq
https://search.emarefa.net/detail/BIM-419944

Modern Language Association (MLA)

al-Ruby, Hamsa Diya Majid. Optical flow estimation based on slantlet transform. (Master's theses Theses and Dissertations Master). University of Technology. (2012).
https://search.emarefa.net/detail/BIM-419944

American Medical Association (AMA)

al-Ruby, Hamsa Diya Majid. (2012). Optical flow estimation based on slantlet transform. (Master's theses Theses and Dissertations Master). University of Technology, Iraq
https://search.emarefa.net/detail/BIM-419944

Language

English

Data Type

Arab Theses

Record ID

BIM-419944