الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 103 |  |  |
| | | from | 103 | | |
Abstract
With the development of multiple types of sensors, huge amount of data have become available for scientific researches. As the volume of data grows, so the need to combine data gathered from different sources to extract the most useful information has become the focus of attention of researchers. Multiple image fusion seeks to combine information from multiple sources to achieve inferences that are not feasible from a single sensor. The purely 3D image-based reconstruction of scene geometry, for instance via a stereo method, is still a highly challenging problem. The primary reason for this is the notorious difficulty of finding multi-view correspondence when visible texture is sparse or complex occlusions are present. Utilizing Time-of-Flight (ToF) camera for 3D image construction faces many challenges, so the current trend in fusion depends on combining different characteristics of different images to build 3D view.
In this thesis, a modified 3D image reconstruction method that uses F- transform is proposed. The suggested model employs shift invariant lossless transform with artificial intelligence using fuzzy set partition to obtain artifact free fused image with clear edges. F-transform helps to combine multiple images into a single 3D one, retaining important features from each, and providing a more accurate description of the object. Ideally, all the meaningful information from the input images will be collected (i.e., the low frequencies and in-plane high frequencies), while discarding the parts bearing no information (i.e., missing high frequencies in the slice-selection direction). The experimental results confirm that the fused images obtained using the proposed model contain richer features and show superior results as compared to DWT fusion. It has the lowest value of RMSE with an average percentage improvement of 16.18%, higher NCC with average percentage of 2.8 %, average enhancement in EM of 7.4%, and average percentage increase in the AG of 6.8%.