Gradus

VOL 3, NO 2 (2016): AUTUMN (NOVEMBER)

 

SZTEREO ALGORITMUS LÉGKÖRI FELHŐK REKONSTRUKCIÓJÁHOZ

STEREO ALGORITHM FOR ATMOSPHERIC CLOUD RECONSTRUCTION


Kátai-Urbán Gábor, Vilem Otte, Megyesi Zoltán, Paul S. Bixel

Abstract

Ebben a cikkben bemutatunk egy légköri felhők felületét rekonstruáló eljárást, ami sztereó kamerarendszer képeit használja fel. Bemutatásra kerül az egész folyamat a javasolt kamerarendszertől a 3D ponthalmaz előállításáig, de a fő hangsúly a felhőszegmentáláson és a sztereó rekonstrukció képfeldolgozási lépésein lesz.

This article presents a method for reconstructing atmospheric cloud surfaces using a stereo camera system. The whole pipeline from the proposed camera system to the creation of the 3D point set is discussed, but the focus is mainly on cloud segmentation and on the image processing steps of stereo reconstruction.


Keywords

Kulcsszavak: Halszemoptika, Sztereo rekonstrukció, Szférikus kameramodell, Sűrű illesztés,

Keywords: Fish-Eye Camera, Stereo Reconstruction, Spherical Camera Model, Dense Matching,


References

[1] Calbó, J.; Sabburg, J.; 2008. Feature extraction from whole-sky ground-based images for cloud-type recognition;Journal of Atmospheric and Oceanic Technology, vol. 25, no. 1, pp. 3-14, 2008.
[2] Dev, S.; Yee Hui Lee ; Winkler, S.; 2014. Systematic study of color spaces and components for the segmentation ofsky/cloud images; Image Processing (ICIP), 2014 IEEE International Conference on , pp. 5102-5106, 2014.
[3] Hartley, R.; Zisserman. A.; 2003. Multiple View Geometry in Computer Vision. Cambridge University Press,Cambridge, UK, pp. 337-360, 2003.
[4] Johnson, R. W. ; Hering, W. S.; Shields, J. E.; 1989. Automated visibility and cloud cover measurements with asolid state imaging system; University of California, San Diego, Scripps Institution of Oceanography, MarinePhysical Laboratory, pp. 59-69, 1989.
[5] Kim, H.; Hilton, A.; 2013. 3D Scene Reconstruction from Multiple Spherical Stereo Pairs; International Journal ofComputer Vision, vol. 104, issue 1, pp. 94-116, 2013.
[6] Li, S.;2008. Binocular Spherical Stereo. IEEE Transactions on Intelligent Transportation Systems, vol. 9, no. 4,2008.
[7] Long, C. N.; Sabburg, J. M.; Calbó, J.; Pag?s D.; 2006. Retrieving cloud characteristics from ground-based daytimecolor all-sky images; Journal of Atmospheric and Oceanic Technology, vol. 23, no. 5, pp. 633–652, 2006.
[8] Megyesi, Z.; Kos, G; Chetverikov, D.; 2006. Dense 3D reconstruction from images by normal aided matching.Machine Graphics & Vision International Journal; vol. 15, issue 1, pp. 3 – 28, 2006.
[9] Mindru, F.; Tuytelaars, T.; Gool, L. V.; Moons, T.; 2004. Moment invariants for recognition under changingviewpoint and illumination. Comput. Vis. Image Underst., vol. 94, issues 1–3, pp. 3–27, 2004.
[10] Salman, N.; Yvinec, M.; 2010. Surface Reconstruction from Multi-View Stereo of Large-Scale Outdoor Scenes..International Journal of Virtual Reality, IPI Press, 2010, The International Journal of Virtual Reality, vol. 9, no. 1, pp.19-26, 2010.
[11] Sara R. Finding the largest unambiguous component of stereo matching. In Proceedings 7th European Conferenceon Computer Vision, Lecture Notes in Computer Science, vol. 3, pp. 900--914, 2002.
[12] Shields, J. E.; Karr, M. E.; Johnson, R. W.; Burden A. R.; 2013. Day/night whole sky imagers for 24-h cloud and skyassessment: history and overview; Applied Optics, vol. 52, issue 8, pp. 1605-1616; 2013.
[13] Sonka, M.; Hlavac, V.; Boyle, R..; 2008. Image Processing, Analysis, and Machine Vision. Thomson-Engineering,pp. 566-573, 2008.



Copyright (c) 2019 Gradus