Image stitching approach using minimum average correlation energy /Siti Salbiah Samsudin

Siti Salbiah , Samsudin (2012) Image stitching approach using minimum average correlation energy /Siti Salbiah Samsudin. Masters thesis, University of Malaya.

PDF (Thesis M.A)
Download (3168Kb) | Preview


    Panorama Image Stitching has been the subject of interest over the decades. A panorama image is the output of blending a set of overlapping images taken at different viewpoints. This process of producing a panorama image is called image stitching, which consists of image registration and image blending. In this thesis, an image stitching method that utilizes Minimum Average Correlation Energy (MACE) filters is used to merge a number of Two sets of analysis were conducted to measure the effectiveness of the proposed method using natural images and radiology medical images. The performance of the proposed method is then compared to those of two correlation based methods namely the Phase Only Correlation (POC) and Normalized Cross Correlation (NCC) methods on the same databases. In the first set of analysis, pair of overlapping and non-overlapping x-ray images of human spine and hand was presented to the system to be merged if they were related or rejected if otherwise. In the second set of analysis involving images of natural scenery, a number of overlapping images were presented to the system to test its ability to recognize the correct matching points to merge them. Then pairs of non-overlapping images were introduced to the system to check whether the system able to identify that if they were not related and should not be merged. In both analyses, it was found that the proposed method outperformed the POC and NCC methods in identifying both the overlapping and non-overlapping images. The efficacy of the proposed method is further vindicated by its average execution time which is about two and five times shorter than those of the POC and NCC method respectively.

    Item Type: Thesis (Masters)
    Additional Information: Thesis (M.Eng.) - Faculty of Engineering, University of Malaya, 2012.
    Uncontrolled Keywords: Panorama Image Stitching; Blending a set; Overlapping images; Average correlation energy
    Subjects: T Technology > T Technology (General)
    T Technology > TA Engineering (General). Civil engineering (General)
    Divisions: Faculty of Engineering
    Depositing User: Mr Prabhakaran Balachandran
    Date Deposited: 18 May 2018 12:22
    Last Modified: 18 May 2018 12:22

    Actions (For repository staff only : Login required)

    View Item