Recognition of characteristic objects on the ground using the SURF method

Authors: Pimenova M.B.
Published in issue: #10(39)/2019
DOI: 10.18698/2541-8009-2019-10-540
Category: Informatics, Computer Engineering and Control | Chapter: System Analysis, Control, and Information Processing, Statistics
Keywords: SURF method, image processing, image recognition, key points, feature point descriptors, camera calibration, machine vision
Published: 18.10.2019

One of the most resistant to projective distortion methods — Speeded Up Robust Features (SURF), which consists in finding specific points of the image. Due to its invariance to shooting conditions and projective transformations, the SURF algorithm can be used to search for objects in real time, solve a navigation problem, and determine the current fragment of the underlying surface in order to obtain the coordinates of the aircraft. The process of recognizing a given fragment on a survey in real time is described. Various image pre-processing algorithms are considered to improve the efficiency of pattern recognition. A study was made of the recognition capabilities in conditions of different illumination, during deformation by scaling and rotation, when changing the level of brightness and blur of the image, as well as the point of view.

References

[1] Bychkov S.S. Klassifikatsiya metodov raspoznavaniya dorozhnykh znakov po videoposledovatel’nosti. Reshetnevskie chteniya, 2017, no. 21-2, pp. 313–314 (in Russ.).

[2] Oleynik A.L. Application of binary descriptors to multiple face tracking in video surveillance systems. Nauchno-tekhnicheskiy vestnik informatsionnykh tekhnologiy, mekhaniki i optiki [Scientific and Technical Journal of Information Technologies, Mechanics and Optics], 2016, vol. 16, no. 4, pp. 670–677 (in Russ.).

[3] Pankov V.V., Kaplieva N.A. Creating panoramic images with the computer vision methods. Vestnik VGU, Seriya: Cistemnyy analiz i informatsionnye tekhnologii [Proceedings of Voronezh State University. Series: Systems analysis and information technologies], 2014, no. 4, pp. 71–74 (in Russ.).

[4] Polovinkin A.N. Algorithms for image classification with a large number of object categories. Vestnik NNGU [Vestnik of Lobachevsky University of Nizhni Novgorod], 2013, no. 4-1, pp. 225–230 (in Russ.).

[5] Pastushkov A.V., Kalayda V.T. [Search methods and algorithms of an object in videostream]. Sbornik nauchnykh trudov SWorld [Proc. SWorld], 2013, vol. 6, no. 3, pp. 38–42 (in Russ.).

[6] Bay H., Ess A., Tuytelaars T., et al. “Speeded-up robust features (SURF)”. Comput. Vis. Image Und., 2008, vol. 110, no. 3, pp. 346–359. DOI: 10.1016/j.cviu.2007.09.014 URL: https://www.sciencedirect.com/science/article/pii/S1077314207001555

[7] Thambi M.S.S.M.S., Menon M.V.R. Offline text document authorization on the basis SIFT and SURF. IJSTE, 2015, vol. 1, no. 10, pp. 328–331.

[8] Lowe D.G. Object recognition from local scale-invariant features. Proc. Int. Conf. Computer Vision, 1999, pp. 1150–1157. DOI: 10.1109/ICCV.1999.790410 URL: https://ieeexplore.ieee.org/document/790410

[9] Dyshlyuk V.O. Study on quality parameters and data points search speed on images using SURF method. Molodoy uchenyy [Young Scientist], 2018, no. 27, pp. 23–26 (in Russ.).

[10] Lowe D.G. Distinctive image features from scale-invariant keypoints. Int. J. Comput. Vis., 2004, vol. 60, no. 2, pp. 91–110. DOI: 10.1023/B:VISI.0000029664.99615.94 URL: https://link.springer.com/article/10.1023%2FB%3AVISI.0000029664.99615.94

[11] Gonzalez R.C., Woods R.E. Digital image processing. Prentice Hall, 2002.

[12] Gruzman I.S., Kirichuk V.S., Kosykh V.P., et al. Tsifrovaya obrabotka izobrazheniy v informatsionnykh sistemakh [Digital image processing in information systems]. Novosibirsk, NGTU Publ., 2002 (in Russ.).

[13] Hartley R., Zisserman A. Multiple view geometry in computer vision. Cambridge university press, 2003 (in Russ.).

[14] Jähne B. Digital image processing. Springer, 2005. (Russ. ed.: Tsifrovaya obrabotka izobrazheniy. Moscow, Tekhnosfera Publ., 2007.)