Please login first
AUTOMATED MEASUREMENT OF PLANT HEIGHT OF WHEAT GENOTYPES USING A DSM DERIVED FROM UAV IMAGERY
* 1 , 1 , 2 , 3
1  Akdeniz University, Faculty of Science, Dept.of Space Science and Technologies
2  Akdeniz University, Faculty of Agriculture, Department of Field Crops
3  Akdeniz University, Institute of Science and Technology, Department of Remote Sensing and GIS

Abstract:

Turkey is a country with a good climate and ecological properties for agricultural production, and the agriculture occupies 24.6% workforce of the whole country. Wheat production is important for Turkish economy and Turkey produced 17 million tons of wheat in 2016.

Traditionally, the monitoring of the wheat length is performed with field works. The breeders and agronomists measure the length of the wheat genotypes with random selection in predefined interval distances. But it is time consuming and not accurate since it is not possible to measure the all wheat genotypes tested on the field experiments. Thus, automated and accurate methods are needed.

High resolution imagery allows producing accurate 3D model of any object including agricultural fields. UAV technology gives an opportunity to acquire imagery from above and then photogrammetric workflow can produce high resolution orthoimage and also 3D model. UAV technology also allows repeating the process in predefined dates to monitor the growth of plant height of wheat genotypes periodically.

In this study, we have evaluated the use of UAV photogrammetry for monitoring of a wheat experiment under field condition, filtered DSM to derive the wheat heights, and compare the results with the field measurements. The images are acquired with use of low cost UAV Walkera QR350 and GoProHero3+ action camera in May 2015. 477 images were acquired. The calculated average wheat height is 77.962 cm and standard deviation is 16.525 cm. In a previous work (Karagöz et al 2017), barley heights are measured as 72.6 cm with 15.2 cm standard deviation with use of traditional methods explained below.

For quality assessment of the proposed method, a reference dataset is collected with terrestrial fieldwork.36 bread wheat genotypes were tested under triple latis design with three replications in Antalya, Turkey. Plot size were 5 m length and 1,2 m width A special circle with 1 m diameter is placed above the plots and the average wheat height in the plot which intersects with the circle was reported as wheat height for the selected plot. For comparison of field measurements with DSM-derived heights, the maximum calculated height in the plot is selected.. The mean, median, standard deviation are calculated as 4.66 cm, 3.75 cm., 13.78 cm. .Regarding statistical T-test between the field measurements and heights from DSM, T-Value is calculated as1,82 and P-Value is 0,071. Since T-value is larger than 0.50, the values between traditional method and our approach are highly correlated while p-value confirms this result.

References:

KARAGÖZ, A., ÖZBEK, K., AKAR, T., ERGÜN, N., AYDOĞAN, S., AYIM, İ, 2017. Agro-Morphological Variation Among an Ancient World Barley Collection, Journal of Agricultural Sciences, Volume 23(4), 444-452.

Keywords: UAV, photogrammetry, wheat, precise agriculture, DSM
Top