
Image Mosaicing for Wide Angle Panorama
Author(s) -
G. Divya,
Chinta Chandra Sekhar
Publication year - 2015
Publication title -
international journal of electrical and computer engineering
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.277
H-Index - 22
ISSN - 2088-8708
DOI - 10.11591/ijece.v5i5.pp1216-1226
Subject(s) - image stitching , ransac , panorama , artificial intelligence , computer vision , computer science , scale invariant feature transform , image registration , feature extraction , feature (linguistics) , pattern recognition (psychology) , image (mathematics) , linguistics , philosophy
Images are integral part in our daily lives. With a normal camera it is not possible to get a wide angle panorama with high resolution. Image Mosaicing is one of the novel techniques, for combining two or more images of the same scene taken in different views into one image. In the dark areas, the obtained image is a panoramic image with high resolution without mask. But in the case of lighting areas, the resultant image is generating mask. In order to gets wide angle panorama, in the existing system, extracting feature points, finding the best stitching line, Cluster Analysis (CA) and Dynamic Programming (DP) methods are used. Also used Weighted Average (WA) method for smooth stitching results and also eliminate intensity seam effectively. In the proposed system, to get feature extraction and feature matching SIFT (Scaled Invariant Feature Transform) algorithm used. In this process, outliers can be generated. RANSAC (Random Sample Consensus) is used for detecting the outliers from the resultant image. Masking is significantly reduced by using Algebraic Reconstruction Techniques (ART).