This paper presents a new image stitching algorithm for mobile multimedia devices. In general, the stitched image obtained by optimal seam finding can avoid the ghost effects but sometimes experience the unconformity of the image structure and color differences near the transition region between images. To solve these problems, the proposed algorithm adopts the seam carving/inserting operator and exploits an adaptive color blending algorithm. Experimental results show that the proposed algorithm can produce a stitched image without visible artifacts including the ghost effects, the unconformity of the structure, and the visible color differences. With respect to the visual quality of the stitched images, the proposed algorithm significantly outperforms simple conventional stitching algorithms and is comparable to the computationally demanding Poisson image editing algorithm 1.
Bibliographical noteFunding Information:
1This research was supported by Mid-career Researcher Program through NRF grant funded by the MEST (No. 2011-0000200) and by Business for Cooperative R&D between Industry, Academy, and Research Institute funded Korea Small and Medium Business Administration in 2011 (Grants No. 00048009).
- content-aware image stitching
- image blending
- optimal seam
- seam carving
- seam inserting
ASJC Scopus subject areas
- Media Technology
- Electrical and Electronic Engineering