Image mosaicking or image stitching is a process of combining two or more images to create a large panoramic image.
The resultant image can also be used for texture mapping of a 3D environment. Consider the case of images taken from a normal
camera. We can convert those images in to mosaicked image using mosaicking technique. This paper presents a new method of image
mosaicking based on SURF and VORONOI. SURF algorithm is used to extract features from the input images. Then the input images
are subdivided in to regions based on VORONOI method. In order to match the regions from VORONOI, ZNCC (Zero Mean
Normalized Cross Correlation) method is used and geometric transformation of input images are calculated based on that value the
two images are merged together to form mosaicked image. Finally to create a natural-looking mosaics, a quasi homography warp is
applied, which balances the perspective distortion against projective distortion in non-overlapping region. This method helps to reduce
projection time and execution time. It is faster than traditional mosaicking algorithms.
Published In:IJCSN Journal Volume 6, Issue 3
Date of Publication : June 2017
Pages : 313-319
Figures :12
Tables : 01
Arya Vijayakumaran Nair : received the B. Tech degree in
Information Technology from M. G University, Kerala, India, in
2012, and currently doing M. Tech in Cyber Security in MET’S
School of Engineering, Mala – APJ Abdul Kalam Technological
University, Kerala, India.
Dr. D. Loganathan : is a Professor and Head of Computer Science
and Engineering department in MET’S School of Engineering,
Mala, Trissur, Kerala. After his B.E., and M.E degree, he
accomplished a doctoral degree from Anna University, Chennai,
India. He has more than 20 years of teaching experience and
having 8 years of research experience in engineering field. His
research interest includes Wireless Communication, Wireless Ad
hoc Networks and Image Processing. He has published several
research papers in various international journals.
Asha. S : is an Assistant professor in MET’S School of Engineering
Mala, Thrissur, Kerala. Received the Btech degree in Govt.
Engineering College Idukki, Kerala and M.Tech from Anna
University. She has more than 4 years of teaching experience.
Subject of interest are programming languages such as C, C++,
Theory of computation, Compiler design.
Image mosaicking, feature extraction, direct methods, SURF, VORONOI, ZNCC
An improved method of image mosaic based on
VORONOI regions and SURF algorithm is proposed here.
In this method the control pints from two input images are
extracted using SURF algorithm. VORONOI method is
used to divide the images in different regions based on the
control points. Then using ZNCC correlation VORONOI
region matching is performed. Finally geometric
transformation is calculated between the two images, for
this the regions with highest correlation scores is selected
and the input images are overlapped to get the desired
mosaicked. To create a natural-looking mosaics, a quasi
homography warp is applied, which balances the
perspective distortion against projective distortion in nonoverlapping
region
[1] C.-C. Lin, S. U. Pankanti, K. N. Ramamurthy, and A. Y.
Aravkin, “Adaptive as-natural-as-possible image
stitching,” in Proc. IEEE Conf. Comput. Vis. Pattern
Recog., Jun. 2015, pp. 1155–1163.
[2] C.-H. Chang, Y. Sato, and Y.-Y. Chuang, “Shapepreserving
halfprojective warps for image stitching,” in
Proc. IEEE Conf. Comput. Vis.Pattern Recog., May
2014, pp. 3254–3261.
[3] D. G Lowe, "Object recognition from local scaleinvariant
features?, International Conference on
Computer Vision, vol. 2, pp.1150– 1157,1999 J. Clerk
Maxwell, A Treatise on Electricity and Magnetism, 3rd
ed., vol. 2. Oxford: Clarendon, 1892, pp.68-73.
[4] D. G. Lowe, ?Distinctive Image Features from Scale-
Invariant Keypoints?, International Journal of Computer
Vision, vol. 60, pp. 91- 110, 2004. M. Young, The
Technical Writer’s Handbook. Mill Valley, CA:
University Science, 1989.
[5] D. G. Lowe, ?Local feature view clustering for 3D
object recognition?, IEEE Conference on Computer
Vision and Pattern Recognition, Hawaii, pp.682-688,
2001. K. Elissa, “Title of paper if known,” unpublished.
[6] D. T. Lee and A. K. Lin. Generalized Delaunay
triangulation for planar graphs. Discrete Comput. Geom.,
1:201–217, 1986.
[7] D. T. Lee. On k-nearest neighbor Voronoi diagrams in
the plane. IEEE Trans. Comput., C-31:478–487, 1982.
[8] Du, Qiang, Vance Faber, and Max Gunzburger.
“Centroidal Voronoi Tessellations: Applications and
Algorithms.” SIAM Review 21 (1999): 637-676.
[9] F. Aurenhammer and H. Edelsbrunner, “An Optimal
Algorithm for Constructing the Weighted Voronoi Diagram in the Plane.” Pattern Recognition 17, 2 (1984):
251-257.
[10] Fang X, Zhu J, Luo B (2012) Image mosaic with relaxed
motion. SIViP 6(4):647–667. doi:10.1007/s11760-010-
0194-4
[11] Fortune, Steven. “A Sweepline Algorithm for Voronoi
Diagrams.” Proceedings of the Second Annual ACM
Symposium on Computational Geometry Yorktown
Heights, New York: Association for Computing
Machinery, 1986: 313-322.
[12] H. Bay, et al., ?Speeded up Robust Features?, Proc. of
the 9th European Conf. on Computer Vision, Cambridge,
U.K., pp. 404-417, 2006
[13] J. Beis, and D. G. Lowe, ?Shape indexing using
approximate nearest-neighbour search in highdimensional
spaces?, Conference on Computer Vision
and Pattern Recognition, Puerto Rico, pp. 1000–1006,
1997..
[14] Ju, Lili, Qiang Du, and Max Gunzburger. “Probabilistic
Methods for Centroidal Voronoi Tessellations and Their
Parallel Implementations.” Parallel Computer 28
(2002):1477-1500.
[15] M. Laraqui, A. Saaidi , A. Mouhib ,M Abarkan(2015)
Images matching using voronoï regions propagation. 3D
Res 6(3):1–16. doi:10.1007/s13319-015-0056-5
[16] M. Lhuillier , L. Quan (2002) Quasi-dense
reconstruction from image sequence. In Computer
Vision—ECCV 2002, 125–139. Springer Berlin
Heidelberg. doi:10.1007/3-540-47967-8_9
[17] N. I. S. with the Global Similarity Prior, “Yu-sheng chen
and yung-yuchuang,” in Proc. Eur. Conf. Comput. Vis.,
Oct. 2016, pp. 186–201.
[18] N.-M. Lˆe. Randomized incremental construction of
simple abstract Voronoi diagrams in 3-space. In Proc.
10th Internat. Conf. Fund. Comput. Theory, volume 965
of Lecture Notes Comput. Sci., pages 333–342. Springer-
Verlag, 1995.
[19] Okabe, Atsuyuki, Barry Boots, Kokichi Sugihara, and
Sung Nok Chiu. Spatial Tessellations: Concepts and
Applications of Voronoi Diagrams, 2nd Ed.. Chichester,
West Sussex, England: John Wiley & Sons, 1999.
[20] P J Anju , Dr.D.Loganathan, (2016) Image Fusion using
NSCT Theory and Wavelet Transform for Medical
Diagnosis, International Journal of Computer Science
and Information Technologies, Vol. 7 (3) , 1507-1510
[21] S. Peleg and J. Herman, ?Panoramic mosaics by manifold
projection? Proc. of IEEE Computer Society conference
on Computer Vision and Pattern Recognition, San Juan,
pp.338-343, 1997.
[22] Wang X, Ying X, Liu Y-J, Xin S-Q,WangW, Gu
X,Mueller-WittigW, He Y (2015) Intrinsic computation
of centroidal Voronoi tessellation (CVT) on meshes.
Comput Des, 51–61. doi:10.1016/j.cad.2014.08.023.