Journal Highly available systems №3 for 2020 г.
Article in number:
Modification of the algorithm for minimizing distortion when stitching aerial photographs into a panorama
Type of article: scientific article
DOI: 10.18127/j20729472-202003-05
UDC: 004.93
Authors:

P.O. Arkhipov – Ph.D.(Eng.), Senior Research Scientist, 

Orel Branch of the FRC «Computer Science and Control» of RAS

E-mail: arpaul@mail.ru

M.V. Tsukanov – Research Engineer, 

Orel Branch of the FRC «Computer Science and Control» of RAS E-mail: tsukanov.m.v@yandex.ru

Abstract:

When stitching large-scale panoramas, a situation often arises when the joints between adjacent frames are well distinguishable and require additional methods of mixing colors in the areas of their overlap. In addition, visible seams appear at the place where the images were stitched, caused by differences in the brightness of the images and the difference in the color pixels themselves. The main methods used to eliminate visually distinguishable boundaries are methods of spatial filtering at the boundaries of stitched images and methods for correcting brightness and color.

Goal of the work is to develop a new modified algorithm for seamless image splicing into a panorama with a reduced number of rewrites of the splicing core. The proposed modified algorithm is focused on stitching panoramas from a large number of images without loss of quality and visible joints.

Solving the problems described in the article made it possible to create a modified algorithm for minimizing distortions when stitching aerial photographs into a panorama, the peculiarity of which is that small intermediate panoramas are created with pre-normalized characteristics of the bonding quality and brightness of images. Intermediate panoramas speed up the process of stitching a single panorama compared to traditional panorama methods.

Pages: 52-57
For citation

Arkhipov P.O., Tsukanov M.V. Modification of the algorithm for minimizing distortion when stitching aerial photographs into a panorama. Highly Available Systems. 2020. V. 16. № 3. P. 52−57. DOI: 10.18127/j20729472-202003-05. (In Russian).

References
  1. Agarwala A. Efficient gradient-domain compositing using quadtrees. ACM Transactions on Graphics. 2007. V. 26. № 3. P. 94.
  2. Burt P.J., Adelson E.H. A multiresolution spline with application to image mosaics. ACM Transactions on Graphics. 1983. V. 2. № 4. P. 217−236.
  3. Farbman Z., Fattal R. and Lischinski D. Convolution pyramids. ACM Trans. Graph. Dec. 2011. V. 30. № 6. P. 175:1−175:8.
  4. Farbman Z., Hoffer G., Lipman Y., Cohen-Or D. and Lischinski D. Coordinates for instant image cloning. ACM Transactions on Graphics. 2009. V. 28. № 3. P. 67.
  5. Perez P., Gangnet M., Blake A. Poisson image editing. ACM Transactions on Graphics (SIGGRAPH’03). 2003. V. 22. № 3. P. 313−318.
  6. Tanaka M., Kamio R., Okutomi M. Seamless image cloning by a closed form solution of a modified Poisson problem. In SIGGRAPH Asia 2012 Posters. SA ’12. New York, NY, USA. 2012. P. 15:1−15:1.
  7. Arkhipov P.O., Sidorkin I.I., Tsukanov M.V. Algoritmicheskaya model tekhnologii minimizatsii iskazhenii pri sshivanii snimkov, poluchennykh s BPLA. Sistemy vysokoi dostupnosti. 2018. № 5. S. 30−35. DOI: 10.18127/j20729472-201805-04. (In Russian).
  8. Arkhipov P.O., Tsukanov M.V. Algoritmicheskaya model ustraneniya otlichii urovnya osveshchennosti mezhdu snimkami, sostavlyayushchimi fotoplan. Informatsionnye sistemy i tekhnologii. 2019. № 4(114). S. 23−29. (In Russian).
  9. Arkhipov P.O., Tsukanov M.V. Algoritmicheskaya model metoda detektirovaniya anomalii pri aerofotosieemke obieektov zhilishchnokommunalnogo khozyaistva. Informatsionnye sistemy i tekhnologii. 2019. № 6(116). S. 39−45. (In Russian).
Date of receipt: 5 июля 2020 г.