الفهرس 
Chapter 1 IntroductionOnly 14 pages are availabe for public view
 |  | from | 123 |  |  |
| | | from | 123 | | |
Abstract
The thesis is concerned with the problem of simultaneous enhanced fusion and interpolation of infrared (IR) and visible images. IR imaging
techniques can be used to image the temperature distribution of an object.
IR images have low contrast between background and targets and small
Signal-to-Noise Ratios (SNRs). These characteristics reduce the
detectability of targets from IR images. In order to recognize targets
correctly from these images, efficient enhancement approaches must be
applied, firstly. The aim of image enhancement is improving the visibility
of low contrast images, decreasing the noise to obtain an image with as
much detail as possible, and to produce reliable methods that represent the
visual information, obtained from a number of imaging techniques, in a
single fused image with more information.
In this thesis, two suggested enhancement approaches for IR images
are presented; a visual quality enhancement approach and a resolution
enhancement approach. The visual quality approach is based on curvelet
transform and homomorphic processing using the additive wavelet
transform (AWT), Different image fusion methods are presented.
The traditional techniques of fusion such as wavelet based fusion are
discussed in the thesis. Then, the idea of curvelet fusion used in satellite
and medical applications is adopted to the fusion of IR and visible images
with success and gives good fusion results. The steps of the curvelet
transform are implemented to both IR and visible images. This transform is
based on the segmentation of the whole image into small overlapping tiles
and then, the ridgelet transform is applied to each tile, then the fusion
process is performed.
II
The resolution enhancement approach is based on image
interpolation, where the enhanced fusion results are then interpolated with
inverse techniques to give high resolution images. Three inverse
interpolation techniques are used for the interpolation of enhanced curvelet
fusion results. The results obtained reveal the superiority of the inverse
interpolation techniques to the polynomial based techniques to obtain high
resolution fusion results.
The thesis comprises five chapters. Chapter 1 presents an
introduction for the low contrast images. It covers the applications of IR
imaging. In addition, this chapter discuses the two suggested enhancement
approaches, image fusion, and interpolation. Chapter 2 covers some
characteristics of IR images. It gives the different spectrum regions that can
be classified as IR ranges. It gives a good understanding of IR images. This
chapter gives also a description of an IR imaging system and its
applications.
Chapter 3 gives a literature review of image fusion concepts. Some
different techniques of image fusion are revised. The chapter concentrates
on discrete wavelet transform (DWT), and dual tree complex wavelet
transform (DT-ı WT) based image fusion, and the mathematical
representation of the curvelet transform. An enhanced fusion technique for
IR and visible images, which is based on the curvelet transform and
homomorphic processing using the AWT, is presented in this chapter.
Moreover, the enhanced fused image based on the curvelet transform is
compared to wavelet transforms (WT) based image fusion results. Finally,
the image quality metrics, which will be used throughout the thesis for the
assessment of the obtained results, are presented.In chapter4, image interpolation for the enhanced fusion results is
studied. Some image interpolation algorithms are introduced taking into
consideration the low resolution (LR) image degradation model. A
comparison study between these algorithms and the traditional cubic spline
interpolation algorithm is presented.
The concluding remarks and the future work are presented in chapter 5.