(48-3) 10 * << * >> * Russian * English * Content * All Issues

Rare plants detection using a YOLOv3 neural network
L.A. Gorodetskaya 1, A.Y. Denisova 1, L.M. Kavelenova 1, V.A. Fedoseev 1

1 Samara National Research University,
443086, Samara, Russia, Moskovskoye Shosse 34

 PDF, 3129 kB

DOI: 10.18287/2412-6179-CO-1405

Pages: 397-405.

Full text of article: Russian language.

Abstract:
Rare plant species restoration (reintroduction) is one of the main biodiversity conservation activities. Reintroduced plants need constant monitoring in order to study features of their development and control the population state. To reduce the human impact on the natural habitat of plants and simplify the monitoring process, we propose the use of automatic analysis of unmanned aerial vehicles (UAVs) data using the Yolov3 neural network. The article discusses neural network parameters for detecting Paeonia Tenuifolia, reintroduced in the Samara region by ecologists from the Department of Ecology, Botany and Nature Conservation of Samara University. The main issue under research is the possibility of training a neural network from peony images collected in an artificial habitat with a subsequent application to images collected in a natural habitat and the possibilities of using multi-temporal data to improve the network training quality. The experiments have shown that training a neural network exclusively using images collected in the natural habitat makes it possible to achieve a probability of correct detection of peonies of 0.93, while using data obtained at different years allows increasing the probability of correct detection to 0.95.

Keywords:
reintroduction, biodiversity, UAV data, neural networks, YOLOv3.

Citation:
Gorodetskaya LA, Denisova AY, Kavelenova LM, Fedoseev VA. Rare plants detection using a YOLOv3 neural network. Computer Optics 2024; 48(3): 397-405. DOI: 10.18287/2412-6179-CO-1405.

Acknowledgements:
This work was funded by the Russian Science Foundation under project No. 23-11-20013.

References:
  1. Malone EW, Perkin JS, Leckie BM, Kulp MA, Hurt CR, Walker DM. Which species, how many, and from where: Integrating habitat suitability, population genomics, and abundance estimates into species reintroduction planning. Global Change Biology 2018; 24(8): 3729-3748. DOI: 10.1111/gcb.14126.
  2. Paeonia Tenuifolia. Russian red book. Source: <https://redbookrf.ru/pion-tonkolistnyy-paeonia-tenuifolia>.
  3. Paeonia tenuifolia (Europe assessment). The IUCN red list of threatened species. Source: <https://www.iucnredlist.org/species/165143/5981869>.
  4. Randin CF, Ashcroft MB, Bolliger J, Cavender-Bares J, Coops NC, Dullinger S, Dirnböck T, Eckert S, Fernández N, Giuliani G, Guisan A, Jetz W, Joost S, Karger D, Lembrechts J, Lenoir J, Luoto M, Morin X, Price B, Rocchini D, Schaepman M, Schmid B, Verburg P, Wilson A, Woodcock P, Yoccoz N, Payne D. Monitoring biodiversity in the Anthropocene using remote sensing in species distribution models. Remote Sens Environ 2020; 239: 111626. DOI: 10.1016/j.rse.2019.111626.
  5. Cao J, Liu K, Zhuo L, Liu L, Zhu Y, Peng L. Combining UAV-based hyperspectral and LiDAR data for mangrove species classification using the rotation forest algorithm. Int J Appl Earth Obs Geoinf 2021; 102: 102414. DOI: 10.1016/j.jag.2021.102414.
  6. Rodriguez JJ, Kuncheva LI, Alonso CJ. Rotation forest: A new classifier ensemble method. IEEE Trans Pattern Anal Mach Intell 2006; 28(10): 1619-1630. DOI: 10.1109/TPAMI.2006.211.
  7. Belcore E, Pittarello M, Lingua AM, Lonati M. Mapping riparian habitats of Natura 2000 network (91E0*, 3240) at individual tree level using UAV multi-temporal and multi-spectral data. Remote Sens 2021; 13(9): 1756. DOI: 10.3390/rs13091756.
  8. Rigatti SJ. Random forest. J Insur Med 2017; 47(1): 31-39. DOI: 10.17849/insm-47-01-31-39.1.
  9. Alvarez-Taboada F, Paredes C, Julián-Pelaz J. Mapping of the invasive species Hakea sericea using unmanned aerial vehicle (UAV) and WorldView-2 imagery and an object-oriented approach. Remote Sens 2017; 9(9): 913. DOI: 10.3390/rs9090913.
  10. Li QS, Wong FKK, Fung T. Assessing the utility of UAV-borne hyperspectral image and photogrammetry derived 3D data for wetland species distribution quick mapping. Int Arch Photogramm Remote Sens Spat Inf Sci 2017; 42: 209. DOI: 10.5194/isprs-archives-XLII-2-W6-209-2017.
  11. Fan Z, Lu J, Gong M, Xie H, Goodman ED. Automatic tobacco plant detection in UAV images via deep neural networks. IEEE J Sel Top Appl Earth Obs Remote Sens 2018; 11(3): 876-887. DOI: 10.1109/JSTARS.2018.2793849.
  12. de Castro AI, Shi Y, Maja JM, Peña JM. UAVs for vegetation monitoring: overview and recent scientific contributions. Remote Sens 2021; 13: 2139. DOI: 10.3390/rs13112139.
  13. Kattenborn T, Eichel J, Fassnacht FE. Convolutional neural networks enable efficient, accurate and fine-grained segmentation of plant species and communities from high-resolution UAV imagery. Sci Rep 2019; 9(1): 17656. DOI: 10.1038/s41598-019-53797-9.
  14. What’s new in YOLO v3? Towards data science. Source: <https://towardsdatascience.com/yolo-v3-object-detection-53fb7d3bfe6b>.
  15. Yolo v4, v3 and v2 for Windows and Linux. Source: <https://github.com/AlexeyAB/darknet>.
  16. Bârliba FC, Bârliba LL, Bârliba C. Methods of acquisition and data processing carried out with DJI Phantom 4 Pro v2.0 UAV equipment. Res J Agric Sci 2020, 52(1): 14-23.

© 2009, IPSI RAS
151, Molodogvardeiskaya str., Samara, 443001, Russia; E-mail: journal@computeroptics.ru ; Tel: +7 (846) 242-41-24 (Executive secretary), +7 (846) 332-56-22 (Issuing editor), Fax: +7 (846) 332-56-20