Computational prediction in the problem of stereo image identification
https://doi.org/10.17586/2226-1494-2024-24-1-11-19
Abstract
The paper examines the issues of increasing the efficiency and reliability of stereo image identification through computational prediction of the position and size of the uncertainty zone in which the desired correspondence pointis known to be located. A control point is selected on one of the stereo images, for which it is necessary to find a correspondence point on the second stereo image. Based on the known parameters of the stereoscopic television system and the coordinates of the control point, using the mathematical apparatus proposed in the work, the coordinates of the boundaries of the uncertainty zone on the second stereo image are calculated. The second point of correspondence is found by the search procedure by comparing identical small areas with centers in the control point on the first stereo image and in the points of the uncertainty zone on the second; the comparison is made according to the criterion of minimum quadratic mismatch of intensities. The necessary a priori information for implementing the method is the maximum heights of the relief displayed on stereo images. The ratios of linear dimensions on a flat relief and on an image formed according to the principle of central projection were obtained. Relationships have been obtained that make it possible to obtain, by calculation, the coordinates of the correspondence points and the stereoscopic mismatch for stereo images of a flat relief. For stereo images of a volumetric relief, calculation formulas are obtained for determining the boundaries of the zone of uncertainty in the second stereo image within which the search for the point of correspondence is carried out. The correctness and performance of the obtained relationships are confirmed by computer modeling. Limiting the size of the search area by means of calculated prediction of the uncertainty zone makes it possible to reduce the computational and time costs of the search procedure. Due to this, the efficiency of identifying stereo image points increases and the likelihood of false identification decreases.
About the Authors
M. V. SamoilenkoRussian Federation
Marina V. Samoilenko — PhD, Associate Professor, Independent Researcher
Moscow
sc 57191194098
V. A. Hachikian
Moscow Aviation Institute (National Research University)
Russian Federation
Russian Federation
Vladimir A. Hachikian — PhD, Associate Professor, Senior Researcher
Moscow, 125993
References
1. Luhmann T., Robson S., Kyle S., Boehm J. Close-Range Photogrammetry and 3D Imaging. Moscow, LENAND Publ., 2018, 704 p.
2. Nazarov A.S. Photogrammetry. 2nd ed., Minsk, TetraSystems Publ., 2010, 400 p. (in Russian)
3. Pepe M., Costantino D., Alfio V.S., Vozza G., Cartellino E. A novel method based on deep learning, GIS and geomatics software for building a 3D city model from VHR satellite stereo imagery. ISPRS International Journal of Geo-Information, 2021, vol. 10, no. 10, pp. 697. https://doi.org/10.3390/ijgi10100697
4. Ma Y., Li Q., Chu L., Zhou Y., Xu C. Real-time detection and spatial localization of insulators for UAV inspection based on binocular stereo vision. Remote Sensing, 2021, vol. 13, no. 2, pp. 230. https:// doi.org/10.3390/rs13020230
5. Ding J., Yan Z., We X. High-accuracy recognition and localization of moving targets in an indoor environment using binocular stereo vision. ISPRS International Journal of Geo-Information, 2021, vol. 10, no. 4, pp. 234. https://doi.org/10.3390/ijgi10040234
6. Albanwan H., Qin R. A comparative study on deep-learning methods for dense image matching of multi-angle and multi-date remote sensing stereo-images. The Photogrammetric Record, 2022, vol. 37, no. 180, pp. 385–409. https://doi.org/10.1111/phor.12430
7. Fan R., Wang H., Cai P., Wu J., Bocus M.J., Qiao L., Liu M. Learning collision-free space detection from stereo images: Homography matrix brings better data augmentation. IEEE/ASME Transactions on Mechatronics, 2022, vol. 27, no. 1, pp. 225–233. https://doi.org/10.1109/tmech.2021.3061077
8. Yang G., Liao Y. An improved binocular stereo matching algorithm based on AANet. Multimedia Tools and Applications, 2023, vol. 82, no. 26, pp. 40987–41003. https://doi.org/10.1007/s11042-023-151836
9. Han Y., Pan C., Cheng Z., Xu Y. A PTV-based feature-point matching algorithm for binocular stereo photogrammetry. Measurement Science and Technology, 2023, vol. 34, no. 12, pp. 125602. https://doi.org/10.1088/1361-6501/acf875
10. Wei H., Meng L. An accurate stereo matching method based on color segments and edges. Pattern Recognition, 2023, vol. 133, pp. 108996. https://doi.org/10.1016/j.patcog.2022.108996
11. Yakasova N.V. Search algorithms for the regular objects in the image. Educational Resources and Technologies, 2016, no. 2, pp. 277–281. (in Russian)
12. Favorskaya M.N., Toupitsyn I.V. Improving robust method of feature points correspondences on stereo images. Mechanics, Control and Informatics, 2012, no. 3(9), pp. 139–144. (in Russian)
13. Chen M., Duan Z., Lan Z., Yi S. Scene reconstruction algorithm for unstructured weak-texture regions based on stereo vision. Applied Sciences, 2023, vol. 13, no. 11, pp. 6407. https://doi.org/10.3390/app13116407
14. Liu C.W., Wang H., Guo S., Bocus M.J., Chen Q., Fan R. Stereo matching: fundamentals, state-of-the-art, and existing challenges. Autonomous Driving Perception: Fundamentals and Applications. Springer Nature Singapore, 2023, pp. 63–100. https://doi.org/10.1007/978-981-99-4287-9_3
15. Favorskaya M.N., Tupitsyn I.V. Hierarchical method of search of corresponding points at stereoimages. The Siberian Aerospace Journal, 2012, no. 1(41), pp. 62–67. (in Russian)
16. Stepanov D.N. Techniques of feature points matching in the problem of uav’s visual navigation. Bulletin of the South Ural State University. Series: Computational Mathematics and Software Engineering, 2015, vol. 4, no. 4, pp. 32–47. https://doi.org/10.14529/cmse150402
17. Goshin Ye.V., Fursov V.A. Conformed identification in corresponding points detection problem. Computer Optics, 2012, vol. 36, no. 1, pp. 131–35. (in Russian)
18. Orlov V.P., Sharikov E.N. Algorithm of finding and classifying special points of object on the basis of Harris’s detector. Journal Nanoindustry, 2017, no. S(74), pp. 171-178. (in Russian)
19. Guk A.P., Altyntsev M.A. Automatic identification of corresponding points for aerial images of forest areas. Vestnik of SSUGT, 2017, vol. 22, no. 4, pp. 68–77. (in Russian)
20. Zagorskii M.Yu., Bogdanov V.L., Garmanov V.V., Koroleva V.P., Ryabov Yu.V. Mathematical model of stereo images and terrain reconstruction algorithm based on it. Sovremennye problemy distantsionnogo zondirovaniya Zemli iz kosmosa, 2015, vol. 12, no. 3, pp. 36–51. (in Russian)
21. Samoilenko M.V. Vector-matrix method for restoration of point spatial coordinates in stereo photography general case. Scientific and Technical Journal of Information Technologies, Mechanics and Optics, 2018, vol. 18, no. 6, pp. 954–960. (in Russian). https://doi.org/10.17586/2226-1494-2018-18-6-954-960
22. Samoilenko M.V. Effect of Intensity asymmetry on stereo image identification error. Measurement Techniques, 2023, vol. 66, no. 5, pp. 311–319. https://doi.org/10.1007/s11018-023-02229-2
Review
For citations:
Samoilenko M.V., Hachikian V.A. Computational prediction in the problem of stereo image identification. Scientific and Technical Journal of Information Technologies, Mechanics and Optics. 2024;24(1):11-19. (In Russ.) https://doi.org/10.17586/2226-1494-2024-24-1-11-19
Views:
10
Email this article 