Color image denosing using 3D wavelet transform

Dissertant

Abbas, Rabab Farhan

Thesis advisor

Abd al-Munim, Matheel Imad al-Din

University

University of Technology

Faculty

-

Department

Computer Sciences Department

University Country

Iraq

Degree

Master

Degree Date

2013

English Abstract

This thesis aims to denoising a noisy color AVI video type, focusing on point-wise degradations, i.e., considering noise in video sequences with Additive Gaussian white noise type using 2-D and 3-D for each of Fast Discrete Wavelet Transform (FDWT) and framelet transform (DDWT) with threshold methods (Hard , Soft and Semisoft thresholds).

The framelet transform has an ability to remove some of the limitations of wavelet transform.

The given result is that 3-D FDWT has a better result than 2-D FDWT and also 3-D framelet transform is better than 2-D FDWT, 3-D FDWT and 2-D framelet transform in eliminating video noises.

Two proposed methods were applied, the first named “Mixing Sub bands and Double Wiener filter Threshold methods” (MSDWTM).

This method applied with doubly local wiener filter and named “MSDWTM Doubly local wiener filter” and then applied with one local and one global wiener filter and named “MSDWTM One local and one global wiener filter”.

The second method named “Translation-Invariant with Wiener filter” (TIW).

These methods are performed with 2-D, 3-D FDWT and 2-D, 3-D framelet transforms and it gives a better result compared with the original methods using database of color AVI videos type.

Main Subjects

Information Technology and Computer Science

Topics

• Digital electronics

• APA
• MLA
• AMA

Language

English

Data Type

Arab Theses

Record ID

BIM-418328