Akademik Veri Yönetim Sistemi
International Congress on Engineering and Life Science, ICELIS 2018, Kastamonu, Türkiye, 26 - 29 Nisan 2018
Abstract: Crop residue management is very important to control soil erosion. Leaving crop residue on the soil surface
before and during planting provides very good cover for the soil. Measuring percent crop residue cover is an essential
step in improving knowledge about the adoption and impact of conservation tillage. This study aimed at determining the
percent crop residue cover using drone and image processing method in wheat field after harvesting. Measurements were
performed during in August-October 2017. Line transect method was used as a reference method to control the results
obtained from drone and ImageJ image processing method. In order to determine the percent crop residue cover using
drone and ImageJ image processing, photographs were taken with a digital camera of the drone from various places of a
field (A, B, C parcels) covered with wheat surface residues. The crop residue covers of A, B and C parcels were ≤ 30%,
30-60% and ≥ 60%, respectively. Wooden frame of size 1x1 m was used to determine sampling areas. During the tests,
the flight height of the drone was 1.5 m. The captured digital photos were transferred to the computer and converted to
JPEG format. Digital photos were analyzed using the ImageJ image processing software. In the line transect method, the
square-shaped areas with a dimension of 10.6 x 10.6 m with a diagonal length of 15 m were marked on field surface. A
measuring tape was then laid across the field surface at an angle of 45 degrees along a 15 m long diagonal. Intersecting
points with crop surface residues at 30 cm intervals along the single edge of the measuring tape were counted. The percent
crop residue cover was calculated by multiplying the number of intersecting points by the total number of slices and
multiplying the value by 100. At the end of study, the mean values of percent crop residue cover using drone and image
processing software ImageJ (Method 1) and line transect method (Method 2) were determined as 30.33% and 30.00% in
A parcel, 64.55% and 57.33% in B parcel, 89.67% and %91.33 in C parcel, respectively. Variance analysis showed that
the differences between the results obtained from Method 1 and Method 2 were statistically insignificant (p>0.05) in all
parcels.
Keywords: Percent crop residue cover, drone, image processing method, ImageJ, line transect method