Selection of homogeneous zones of agricultural field for laying of experiments using unmanned aerial vehicle

V. M. Bure ; E. P. Mitrofanov ; O. A. Mitrofanova ; A. F. Petrushin

Vestnik Sankt-Peterburgskogo universiteta. Prikladnaâ matematika, informatika, processy upravleniâ, Tome 14 (2018) no. 2, pp. 145-150

Résumé

An important stage in research aimed at solving the problems of accurate farming is the laying of field experiments. A necessary condition for carrying out such experiments is to ensure homogeneity of the selected land plot. Most of the existing techniques for isolating homogeneous zones for conducting experiments are based on costly and labor-intensive sampling and analysis of soil samples. An alternative and promising approach is the use of an unmanned aerial vehicle. In work, all stages of choosing a homogeneous land plot with the help of aerial photography are presented in sufficient detail. The object of the study was a field with a long-term sowing of “goat” on the basis of the Menkovsky branch of the Agrophysical Institute (Leningrad region). Aerial photography was carried out in 2015–2017 with the help of an unmanned aerial vehicle “Geoscan 401”. The received data were processed with the help of specialized software: crosslinking and alignment were carried out in the Agisoft PhotoScan program; the thematic processing and allocation of homogeneous areas of the field were carried out in the programs QGis and Saga Gis. To assess the state of vegetation, the vegetative index NDVI (Normalized Difference Vegetation Index) was applied. To cluster the homogeneous parts of the field in terms of NDVI parameters the ISODATA algorithm (Iterative Self-Organizing Data Analysis Technique Algorithm) was applied. The paper presents the results of clustering images of the same field in different time periods. In the course of the work, intersections of these aerial photographs were constructed, four clusters were identified, which are the intersection of the corresponding homogeneous zones for the considered time periods. Accordingly, the laying of experiments is expedient to be carried out on these sites, since the homogeneity present there seems more stable in dynamics.

Détail

Keywords: aerial photography, precision agriculture, clustering, ISODATA algorithm.

Bibliographie

[1] Dospekhov B. A., Method of field experiment, Kolos Publ., M., 1979, 416 pp. (In Russian)

[2] Bure V. M., Mitrofanova O. A., “Prediction of the spatial distribution of ecological data using kriging and binary regression”, Vestnik of Saint Petersburg University. Series 10. Applied Mathematics. Computer Science. Control Processes, 2016, no. 3, 97–105 (In Russian)

[3] Petrushin A. F., Mitrofanov E. P., Mitrofanova O. A., “Digital terrain model for monitoring reclamation facilities”, Agroecosystems in natural and regulated conditions: from the theoretical model to the practice of precision control, Agrophysical Institute Publ., Saint Petersburg, 2016, 47–51 (In Russian)

[4] Bure V. M., Mitrofanova O. A., “Analysis of aerial photographs to predict the spatial distribution of ecological data”, Contemporary Engineering Sciences, 10:4 (2017), 157–163 | DOI

[5] Iakushev V. P., Kanash E. V., Konev A. A., Kovtiukh S. N., Lekomtsev P. V., Matveenko D. A., Petrushin A. F., Iakushev V. V., Bure V. M., Rusakov D. V., Osipov Iu. A., Theoretical and methodological foundations for the separation of homogeneous technological zones for the differentiated application of chemicalization means based on the optical characteristics of seeding, practical allowance, Agrophysical Institute Publ., Saint Petersburg, 2010, 60 pp. (In Russian)

[6] Tou J. T., Gonzalez R. C., Pattern recognition principles, Addison-Wesley Publ. Company, Boston, MA, USA, 1974, 395 pp. | MR | Zbl