International Journal of Computer Applications |
Foundation of Computer Science (FCS), NY, USA |
Volume 73 - Number 5 |
Year of Publication: 2013 |
Authors: Kawther Abbas Sallal, Abdul Monem Saleh Rahma |
10.5120/12737-9614 |
Kawther Abbas Sallal, Abdul Monem Saleh Rahma . Generation of Video Panorama System. International Journal of Computer Applications. 73, 5 ( July 2013), 20-26. DOI=10.5120/12737-9614
The construction of large, high-resolution view is an active area of research in many fields like computer vision and computer graphics. A Panorama is the process of combining multiple images or videos with overlapping region to produce a panoramic view. it enables producing a complete view of an area that cannot fit in a single shot. This paper describes simple method for taking two or more videos and creating a panoramic video. Two steps of motion estimation is done. The first, estimates the motion between the frames of the two videos and the second estimates the motion to produce the final panoramic video. The process to generate a panoramic view can be divided into four main components: video acquisition, similar regions detection, motion estimation and merging. Geometric moment invariant produces a set of features that are invariant under shifting, scaling and rotation. it is widely used to extract the global features for pattern recognition due to its discrimination power and robustness. In this paper, moment invariant is used to determine the locations of merging. The experimental results are done on videos taken by two horizontally adjacent cameras. The results show that the proposed algorithm is fast and efficient.