Stable descriptors in image recognition tasks
https://doi.org/10.21822/2073-6185-2020-47-3-93-100
Abstract
Objective. The objective of the study is to determine various stable characteristics of images (semi-invariants and invariants) as descriptors necessary for the formation of a feature space of standards intended for recognizing images of different nature belonging to different classes of objects. Methods. The authors propose metrics for evaluating the proximity of the recognized image to a given standard in the space of covariance matrices, based on the obtained descriptors as a methodological basis for constructing image recognition methods. Results. The content of the main stages of selecting descriptors for a given class of objects is developed, taking into account the different illumination of the recognized images. The effectiveness of the results obtained is confirmed by experimental studies related to the solution of the problem of recognition of special images - facies. Conclusions. The definition of stable image descriptors as invariants or semi-invariants to zoom and brightness transformations allows solving the problems of facies classification in conditions of the unstable shooting of recognized images. The images can be rotated and shifted in any way. In general, the proposed approach allows developing an effective image recognition system in the presence of various types of interference on the recognized images.
Supporting agencies: This work was carried out with partial financial support from the Russian Foundation for Basic Research (projects 20-07-00022 A and 17-29-07002 ofi_m).
About the Authors
V. B. Melekhin
Daghestan State Technical University
Russian Federation
Russian Federation
Vladimir B. Melekhin - Dr. Sci. (Technical), Prof., Department of Computer Software and Automated Systems.
70 I. Shamil Ave., Makhachkala 367026.
V. M. Khachumov
A.K. Aylamazyan Institute of Software Systems Russian Academy of Sciences; Federal Research Center "Informatics and Control" Russian Academy of Sciences; Russian Peoples' Friendship University
Russian Federation
Russian Federation
Vyacheslav M. Khachumov - Dr. Sci. (Technical), Prof., Head of the Intelligent Control Laboratory.
4a Petra Pervogo St., Yaroslavl region, Pereslavsky district, Veskovo village 152021; 44 Vavilova St., building 2, Moscow 119333; 6 Miklukho-Maklaya St., Moscow 117198.
References
1. Vinberg E.B., Popov V.L. Teoriya invariantov [Theory of invariants]. Itogi nauki i tekhn. Ser. Sovrem. probl. mat. Fundam. napravleniya, 1989, Vol. 55. рp.137 - 309. (In Russ)
2. D'yodonne ZH., Kerrol Dzh., Mamford D. Geometricheskaya teoriya invariantov [Geometric theory of invariants]. Moscow: Mir, 1974. 278 p. (In Russ)
3. Zubkov A.N. Obzor po teorii invariantov i ee prilozheniyam [Review of the theory of invariants and its applications]. Prikladnaya matematika i fundamental'naya informatika [Applied Mathematics and Fundamental Informatics]. 2014. №1. pp. 45 - 49. (In Russ)
4. Abramov N.S., Hachumov V.M. Raspoznavanie na osnove invariantnyh momentov [Recognition based on invariant moments]. Vestnik Rossijskogo universiteta druzhby narodov. Seriya: Matematika, informatika, fizika [Bulletin of the Peoples' Friendship University of Russia. Series: Mathematics, computer science, physics]. 2014. No. 2. pp.142 - 149. (In Russ)
5. Abramov N.S., Fralenko V.P. Opredelenie rasstoyanij na osnove sistemy tekhnicheskogo zreniya i metoda invari-antnyh momentov [Determination of distances based on the technical vision system and the method of invariant moments]. Informacionnye tekhnologii i vychislitel'nye sistemy [Information technologies and computational systems]. 2014. No.4. pp.32 - 39. (In Russ)
6. Hachumov M.V. Invariantnye momenty i metriki v zadachah raspoznavaniya graficheskih obrazov [Invariant moments and metrics in graphic image recognition problems]. Sovremennye naukoemkie tekhnologii [Modern scienceintensive technologies]. 2020. No.4, Ch.1. pp. 69 - 77. (In Russ)
7. Trushkov V.V., Hachumov V.M. Opredelenie orientacii ob"ektov v trekhmernom prostranstve [Determination of the orientation of objects in three-dimensional space]. Avtometriya [Avtometriya]. 2008. No.3 (44). pp. 75-79. (In Russ)
8. Metody komp'yuternoj obrabotki izobrazhenij [Methods of Computer Image Processing]. Pod red. V.A. Sojfer . Moscow: Fizmatlit, 2003. 784 p. (In Russ)
9. Pavlidis T. Algoritmy mashinnoj grafiki i obrabotki izobrazhenij [Algorithms of computer graphics and image processing]. Moscow: Radio i svyaz', 1991. 400 p. (In Russ)
10. Mazhuga V.V., Hachumov M.V. Algoritmy obrabotki izobrazhenij dlya klassifikacii sostoyanij biologicheskih sistem [Image processing algorithms for classifying the states of biological systems]. Informacionnye tekhnologii i vychis-litel'nye sistemy [Information technologies and computing systems]. 2012. No.2. pp.54 -63. (In Russ)
11. NTSC. http://en.wikipedia.org/wiki/NTSC
12. Hachumov M.V. Rasstoyaniya, metriki i klasternyj analiz [Distances, metrics and cluster analysis]. Iskusstven-nyj intellekt i prinyatie reshenij [Artificial intelligence and decision making]. No. 1, 2012, pp. 81 - 89. (In Russ)
13. Hachumov M.V. Primenenie nejrona i rasstoyaniya Evklida-Mahalanobisa v zadache binarnoj klassifikacii [Application of a neuron and Euclidean-Mahalanobis distance in the problem of binary classification]. Nauka i sovremennost' [Science and Modernity]. 2010. No. 2-3. pp. 82 - 86. (In Russ)
14. Forstner V., Moonen B. A metric for covariance matrices. Technical report, Dep. Of Geodesy and Geoinformatics, Stuttgart, 1999. pp. 113 - 128.
Review
For citations:
Melekhin V.B., Khachumov V.M. Stable descriptors in image recognition tasks. Herald of Dagestan State Technical University. Technical Sciences. 2020;47(3):93-100. (In Russ.) https://doi.org/10.21822/2073-6185-2020-47-3-93-100
Views:
575
Email this article 
