Marker images detection algorithm for the unmanned aerial vehicle vertical landing

DOI: 10.34759/trd-2021-116-13


Trusfus M. V.*, Abdullin I. N.**

Kazan National Research Technical University named after A.N. Tupolev, KNRTU-KAI, 10, Karl Marks str., Kazan, 420111, Russia



The primary task while drones application consists in automatic landing ensuring. Most of the existing auto-landing solutions ensure accuracy up to three meters. For a more accurate landing, accurate distance measurements are required. Methods, ensuring accurate camera images, require accurate identification of the marker image points.

The marker image consists of two black rectangles situated in the center of each other at the angle of 90 degrees. The rectangles are applied on a white surface. Their detection is based on the objects’ contours detecting on a monochrome image with further test for compliance to marker geometric shape. The dependencies of distances between the points of the geometric shape equivalent to the claimed structure are presented.

This study considers the methods of adaptive binarization and the Canny boundary detector to perform camera image conversion to monochrome one. Detection of the outline of each object is being performed on the monochrome image. A convex hull of minimum area is being built for each outline in such a way that all the object points be inside. The search for the points corresponding to the marker’s apexes is being performed according to the presented dependence of distances on the convex hull. Checking of all points of the initial outline on the correspondence to the marker’s geometry shape by checking if each point lays on the corresponding side is being performed for the marker image identification.

Software was employed for conducting experimental study of the described algorithm. One of the programs employs binarization method, while Canny boundary detector is used in the other one. Photographs in 640 ´ 480, 800 ´ 600 and 1920 ´ 1080 resolution were used while conducting the study.

According to the results of the study, the highest detection accuracy and speed were displayed by an algorithm based on the use of the Canny boundary detector. False detection of marker images was not detected in any of the algorithms. The algorithm based on the adaptive binarization revealed the best performance in detecting marker images in «blurry» images. A conclusion was made by the results of the study that the developed algorithm allows detecting all points of the marker image of the vertical landing site designation without errors in false targets detecting.


marker image, pattern recognition, image processing, vertical landing


  1. Gautam A, Sujit PB, Saripalli S. A survey of autonomous landing techniques for UAVs, 2014 International Conference on Unmanned Aircraft Systems, (ICUAS) 2014, Conference Proceedings, IEEE Computer Society, 2014, pp. 1210 - 1218. DOI: 10.1109/ICUAS.2014.6842377

  2. Fedorenko R.V., Abdullin I.N., Klimchik A.S. Trudy tsentra kompetentsii NTI “Tekhnologii komponentov robototekhniki i mekhatroniki”, Izhevsk, Izhevskii institut komp'yuternykh issledovanii, 2019, pp. 129 – 166.

  3. Aref'ev R.O., Aref'eva N.G., Skrypnik O.N. Trudy MAI, 2017, no. 92. URL:

  4. Pogosyan M.A., Vereikin A.A. Trudy MAI, 2020, no. 113. URL: DOI: 10.34759/trd-2020-113-11

  5. Skryabin Yu.M., Potekhin D.S. Trudy MAI, 2019, no. 106. URL:

  6. Gogolev A.A., Gorobinskii M.A. Trudy MAI, 2018, no. 101. URL:

  7. Trusfus M.V., Eminov F.I., Abdullin I.N. I Vserossiiskaya nauchno-prakticheskaya konferentsiya molodykh aviatorov Rossii “Aviatsiya budushchego: trendy, vyzovy i vozmozhnosti”, Kazan', Izd-vo KNITU-KAI, 2019, pp. 576 - 579.

  8. Altwaijry H., Trulls E., Hays J., Fua P., Belongie S. Learning to Match Aerial Images with Deep Attentive Architectures, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, 2016, pp. 3539 – 3547. DOI: 10.1109/CVPR.2016.385

  9. Korneev M.A., Maksimov A.N., Maksimov N.A. Trudy MAI, 2012, no. 58. URL:

  10. Simo-Serra E. et al. Discriminative learning of deep convolutional feature point descriptors, 2015 International Conference on Computer Vision, ICCV 2015, Institute of Electrical and Electronics Engineers Inc., 2015, pp. 118 - 126. URL:

  11. Chen L.C. et al. Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFs, Computer Science, 2014. URL:

  12. Suzuki S. Topological structural analysis of digitized binary images by border following, Computer Vision, Graphics and Image Processing (ICVGIP), 1985, vol. 30 (1), pp. 32 – 46.

  13. Burger W., Burge M.J. Principles of Digital Image Processing, Springer, London, 2009, DOI: 10.1007/978-1-84800-195-4

  14. Otsu N. A Threshold Selection Method from Gray-Level Histograms, IEEE Transactions on Systems, Man and Cybernetics, 1979, vol. 9, no. 1, pp. 62 - 66. DOI: 10.1109/TSMC.1979.4310076

  15. Dey Nilanjan, Dutta Saurab, Dey Goutami, Chakraborty Sayan, Ray Ruben, Roy Payel. Adaptive thresholding: A comparative study, 2014 International Conference on Control, Instrumentation, Communication and Computational Technologies, ICCICCT 2014. DOI: 10.1109/ICCICCT.2014.6993140

  16. Canny J. A Computational Approach to Edge Detection, IEEE Transactions on Pattern Analysis and Machine Intelligence, 1986, vol. PAMI-8, no. 6, pp. 679 – 69. DOI: 10.1109/TPAMI.1986.4767851

  17. Sklansky J. Finding the convex hull of a simple polygon, Pattern Recognition Letters, 1982, vol. 1, no. 2, pp. 79 – 83. DOI: 10.1016/0167-8655(82)90016-2

  18. Anton H. Elementary Linear Algebra, New York, Wiley, 1993, 280 p.

  19. Lebedev G.N., Mirzoyan L.A., Efimov A.V. Trudy MAI, 2011, no. 48. URL:

  20. Dmitriev V.I., Zvonarev V.V., Lisitsyn Yu.E. Trudy MAI, 2020, no. 112. URL: DOI: 10.34759/trd-2020-112-16

Download — informational site MAI

Copyright © 2000-2022 by MAI