ESTIMATION OF THE VEGETATION COVERAGE FRACTION IN CORN (Zea mays) THROUGH DIGITAL IMAGES TAKEN BY AN UNMANNED AERIAL VEHICLE (UAV)

Main Article Content

Héctor García-Martínez
Héctor Flores-Magdaleno
Abdul Khalil-Gardezi
Roberto Ascencio-Hernández
Leonardo Tijerina-Chávez
Mario A. Vázquez-Peña
Oscar R. Mancilla-Villa

Abstract

The vegetation cover fraction is an important variable in crop monitoring that is related to biophysical characteristics such as density, phenology, leaf area index, germination, photosynthetic capacity of foliage, evapotranspiration, productivity and crop yield. The aim of this study was to estimate the vegetation cover fraction during the development of the maize crop, from images taken by a sensor that captures information in the spectrum of red, green, and blue (RGB), mounted on an unmanned aerial vehicle, with vegetation indices and thresholding methods. Six flights were made, which generated six orthomosaics during the development of the crop. The image capture was 30 m above ground level, with resolutions of 0.75 to 0.80 cm/pixel. For the extraction of the vegetation and transformation of the orthomosaics to grayscale, four vegetation indices (IV): triangular greenness index (TGI), excess green index (ExG), visible atmospheric resistance index (VARI) and normalized green-red difference index (NGRDI) were used. Thresholding was performed with the Otsu, IsoData, Fuzzy, and momentum conservation algorithms for each grayscale image generated with the IVs to separate and classify the pixels of the images into two classes, soil and vegetation. The indices used to extract the vegetation showed high precision in the early stages of development; thus, it is possible to use them to know the germination, density, and early vigor of the crop. It was found that the EXG index presented errors from 2.2 to 17.8 % in the estimation of the vegetation cover fraction during the development of the crop; errors less than 5 % were obtained through the threshold calculated with the Fuzzy method up to 58 days after planting. The use of high-resolution digital images from the unmanned aerial vehicle (UAV) allowed estimating the vegetation cover with errors of less than 5 % in the early stages of the development of the maize crop.

Article Details

Section
Scientific Articles

Most read articles by the same author(s)