Ciência Rural, Santa Maria, v.

51:8,
Evaluation e20200692,
of the 2021
nutritional status of corn by vegetation indices via aerial images.
http://doi.org/10.1590/0103-8478cr20200692
1
ISSNe 1678-4596
soil science

Evaluation of the nutritional status of corn by vegetation indices via aerial images

Aderson Soares de Andrade Junior1* Francisco de Brito Melo1 Edson Alves Bastos1
Milton José Cardoso1
1
Embrapa Meio-Norte, 64008-780, Teresina, PI, Brasil. E-mail: [email protected]. *Corresponding author.

ABSTRACT: The objective of this study is to determine the vegetation indices (IV) as a means of identifying the nutritional status of corn,
with respect to the soil nitrogen and potassium, using the aerial images received through an RGB camera loaded on an unmanned aerial
vehicle. The images were obtained for an experiment of the nitrogen levels (0, 60, 120 and 180 kg ha-1) and potassium levels (0, 50, 100 and
150 kg ha-1), in the random block design, with a factorial scheme of 4 x 4, having three repetitions. Ten leaves were plucked per plot during the
flowering phase to assess the total N (NF) and K+ leaf contents. The Pearson’s correlation analysis, as well as the analyses of variance and
regression between the IV and the concentrations of N and K2O. NF, K+ and the grain yield, responded only to the soil N levels. A significant
correlation was observed for the indices of Red Index, Normalized Difference Index and Visible Atmospherically Resistant Index with the NF,
which endorses them as favorable in identifying the nutritional standing of corn, with respect to the N level. Not even a single one of the indices
evaluated could detect the nutritional ranking of corn in the context of the potassium level.
Key words: Zea mays L, RGB images, remote sensing, precision agriculture.

Avaliação do estado nutricional do milho por índices de vegetação de imagens aéreas

RESUMO: O estudo teve como objetivo avaliar índices de vegetação (IV) para detecção do status nutricional do milho com relação ao nitro-
gênio e potássio por meio de imagens aéreas obtidas por câmera RGB embarcada em veículo aéreo não tripulado. As imagens foram adqui-
ridas em ensaio de níveis de nitrogênio (0, 60, 120 e 180 kg ha-1) e potássio (0, 50, 100 e 150 kg ha-1), em blocos ao acaso, fatorial 4 x 4, com
três repetições. Coletaram-se dez folhas por parcela na fase de florescimento para avaliação do teor foliar de N total (NF) e K+. Efetuou-se
análise de correlação de Pearson, análise de variância e de regressão entre os IV e os níveis de N e de K2O. NF, K+ e a produtividade de grãos
responderam apenas aos níveis de N no solo. Houve correlação significativa para os índices Excess Red Index, Normalized Difference Index e
Visible Atmospherically Resistant Index com o NF, que os credencia como promissores na detecção do status nutricional do milho em relação
ao N. Nenhum dos índices avaliados foi capaz de detectar o status nutricional do milho com relação ao potássio.
Palavras-chave: Zea mays L, imagens RGB, sensoriamento remoto, agricultura de precisão.

INTRODUCTION significance due to the high yield of the corn cultivars

in response to the application of a combination of
For corn to express its productive potential potassium and nitrogen (PETTER et al., 2016).
there is a high nutritional demand where nitrogen From the findings of several studies, the
is the nutrient needed in large amounts. Nitrogen positive and crucial effect of applying nitrogen
exerts the greatest influence on grain production in and potassium on corn grain production is clearly
corn and is the major factor that contributes to the evident (BASTOS et al., 2005; MELO et al., 2011;
expenditure involved in its production (MELO et PETTER et al., 2016). However, the paucity of
al., 2011). studies continues to hinder an accurate evaluation of
The next most important element absorbed the nutritional status of the commercial corn crops,
in high quantities by the corn crop is potassium in terms of the rational and efficient addition of these
(BASTOS et al., 2005). Adding potassium during the nutrients, based on the plant requirement and levels
commercial production of the corn crop is gaining of soil fertility in the regions of production.
Received 07.23.20 Approved 09.29.20 Returned by the author 03.05.21
CR-2020-0692.R1 Ciência Rural, v.51, n.8, 2021.
2 Junior et al.

Proximal remote sensing is an excellent texture (SANTOS et al., 2018). When the chemical
and viable tool which enables the nutritional status and physical characterization of the soil was done
of agricultural crops to be assessed (LI et al., 2014; initially, in the 0.0 - 0.2 m layer, the following
CILIA et al., 2014). The proximal detection method attributes were recorded: pH in H2O of 5.9; pH in
is effective in providing possible automation and CaCl2 of 5.0; 17.2 mmolc dm-3 of potential acidity
mechanization applications, like aerial images of (H+ Al); 0.65 mmolc dm-3 exchangeable aluminum
the crops, employing unmanned aerial vehicles content; 2.0 mmolc potassium dm-3 (K+); 11.9 mmolc
(UAVs). This enables a substantial reduction in dm-3 of calcium (Ca+2), 4.1 mmolc dm-3 of magnesium
the field operation-related costs. The use of UAVs (Mg+2); 11.6 g kg-1 of carbon (C); 18.8 mg dm-3 of
provides results in terms of the spatial resolution of phosphorus (P) (Mehlich); 35 mmolc dm-3 of cation
the images, flexible revisiting time, as well as gives exchange capacity (CTC); 50.5% base saturation
high versatility even when the climatic conditions are (V%); 831.9 g kg-1 of the sand fraction; 46.9 g kg-1 of
unfavorable (TORRES-SANCHEZ et al., 2015) the silt fraction; 121.2 g kg-1 of the clay fraction and
Using the spectral reflectance of the 1.65 g cm-3 of soil density. The analyses done adopted
canopy, the vegetation indices (IV) are simple and the recommendations of the Embrapa Manual of the
successful algorithms, helpful in the quantitative and analysis of soil, plants and fertilizers (SILVA, 2009).
qualitative evaluation of the vegetation cover, as well In the experiment performed regarding the
as plant vigor and growth dynamics (GITELSON nitrogen (N) and potassium (K2O) levels in corn, aerial
et al., 2002). The remote sensing of the vegetation, images were acquired (Figure 1). The experiment was
principally connected with acquiring multispectral conducted from February to June 2019, in a rainfed
images, can be practically applied. Hence, many IVs regime, adopting a randomized block design with
have been suggested using multispectral images for treatments done in a 4 x 4 factorial scheme (N levels
remotely estimating the nutritional status of different versus K2O levels), including three replications (T1: 0
agricultural crops, particularly from the perspective N –0 K2O; T2: 0 N - 50 K2O; T3: 0 N - 100 K2O; T4:
of the soil nitrogen availability (ISLA et al., 2011; 0 N - 150 K2O; T5: 60 N - 0 K2O; T6: 60 N - 50 K2O;
LI et al., 2014; CILIA et al., 2014; VERGARA- T7: 60 N - 100 K2O; T8: 60 N - 150 K2O; T9: 120 N - 0
DÍAZ et al., 2016). However, studies are still in the K2O; T10: 120 N - 50 K2O; T11: 120 N - 100 K2O; T12:
nascent stages in determining the nutritional status 120 N - 150 K2O; T13: 180 N - 0 K2O; T14: 180 N - 50
of potassium (SRIDEVY et al., 2018), and the use of K2O; T15: 180 N - 100 K2O and T16: 180 N - 150 K2O).
RGB (red-green-blue) images received from the less- The experimental plot, extending across an area of 16.0
​​
expensive cameras (RASMUSSEN et al., 2016). m2, included four rows of plants, each row of 8 m length,
Thus, this research was performed to and the two central lines used for the evaluations (​​8.0
remotely assess the nutritional status of the corn m2 of useful area). The spacing between the plant rows
crop, with regard to the soil nitrogen and potassium, was 0.5 m, and plant density was 5 per meter. The corn
using the vegetation indices received from the aerial variety used was the hybrid commercial corn Pioneer
images acquired from an RGB camera, loaded on an 30F35VYHR, sown on 02/13/2019.
unmanned aerial vehicle (UAV). First, sowing was performed in parallel
furrows at 0.15 m depth and with 0.10 m distance
MATERIALS AND METHODS between the sowing lines, manually applying
phosphorus and zinc as fertilizers, using 80 kg ha-1
The current study was done in a plot in of P2O5 and 3 kg ha-1 of Zn, in the forms of triple
Fazenda Weisul Agrícola, in Magalhães de Almeida, superphosphate (TSP) and zinc sulfate, respectively.
MA, with the coordinates of 3° 22’9.27’’ S and 42° 17’ Fertilization with nitrogen and potassium was
28.8’’ W at 85 m altitude. The local climatic conditions accomplished by applying half the quantity of N
are hot, and sub-humid, with moderate water surplus and K2O prescribed for each treatment , at the time
in summer (Aw), annual average temperature of 26 of sowing, and the remainder was added in cover,
ºC, and annual precipitation of 1,250 mm, particularly performed via haul, at a distance of 0.10 m from the
between February and May (CORREIA FILHO et al., planting line, using moist soil. This was performed
2011). During the experimental period, in the area during the late afternoon, at the time of the opening
under study, the total precipitation was 1,010 mm, of the 6th leaf. Urea and potassium chloride were
recorded using a rain gauge. employed, respectively, as the N and K2O sources.
In the experimental area the soil is of the During the flowering time, ten corn leaves
Latossolo Amarelo Distrocoeso type and of medium were collected at random per plot to assess the total

Ciência Rural, v.51, n.8, 2021.

 Evaluation of the nutritional status of corn by vegetation indices via aerial images. 3

Figure 1 - Location of the experimental area. A: Magalhães de Almeida, MA, Brazil; B: Weisul Farm and C: Aerial
image of the experimental area with the disposition of the evaluated treatments.

nitrogen and potassium contents of the leaf, using the This operates in the visible region (red: 660-670
central third of the base leaf of the corn ear, during nm, green: 550-560 nm, blue: 470-480 nm), with f /
the planting stage (50% of the plants in the plot were 2.8 aperture, 3.6 mm focal length, and 4000 x 3000
showing the tassel). The morning (between 8 - 11 am) pixels resolution. The following configuration was
was the best time to collect the leaves on the same used at the time of the flight: ISO 100, opening speed
day as the flight. To evaluate the N and K+ content 1/800 s and white balance of zero. On the day the
present in the leaves, the semi-micro Kjeldahl method leaves were collected between 11:00 am and 12:00
(SILVA, 2009) was followed. On 06/26/2019 the corn am and analyzed for N and K+ analysis (04/09/2019),
was manually harvested, and the dry grain yield was the aircraft was at altitude of 30 m and at speed of
determined at 13% humidity. 2.5 m/s, with camera angle at 90°. The planning and
To receive the aerial images, a quad-type operation of the flight was done in Pix4D-Capture®
UAV, DJI brand, model Phantom 3 Professional, was software (www.pix4d.com). The configuration of the
used, which was provided with a DJI sensor model camera settings was done using the DJI GO® software
FC300X, (DJI, Nanshan District, Shenzhen, China). (www.dji.com).

Ciência Rural, v.51, n.8, 2021.

4 Junior et al.

To ensure the high-quality creation of The orthomosaic was classified under the
the orthomosaic, 80% lateral and frontal overlap supervision of the Gaussian Mixture Model method
was employed during the flight, providing 30 suggested by LAGRANGE et al. (2017). The plugin
aerial photographs in total, to encompass the whole enabled the mosaic to be to be rasterized into two classes
experimental region, with 1.5 cm pixel GSD (ground (soil and leaves). This facilitated the pixels classified as
sample distance). Processing of the orthomosaic soil to be removed from the mosaic, confirming that the
of the aerial images was done using the WEB- estimation of the vegetation index was done using only
OpenDroneMap® software (www.opendronemap. the pixels classified as leaves. This was accomplished
org) beta version 0.3.1. The standard configuration using the QGIS v “dzetsaka” plugin. 2.18 (QGIS, 2016).
of the software enabled a high spatial orthomosaic Evaluation was done of 18 vegetation
resolution (2.5 cm / pixel) to be generated. indices (IV) (Table 1) estimated from the

Table 1 - RGB vegetation indices evaluated in the study.

Indices Sigla Equation References

𝑟𝑟 − 𝑏𝑏
Coloration Index CI = MANDAL (2016)
𝑟𝑟
= 18.78745 + (0.44 r) – (0.88 g)
Color Index of Vegetation Extraction CIVE YANG et al. (2015)
+ (0.385 b)
1 1
Carotenoid Reflectance Index 1 CRI-1 =� �−� � GITELSON et al. (2002)
𝑏𝑏 𝑔𝑔
1 1
Carotenoid Reflectance Index 2 CRI-2 =� �−� � GITELSON et al. (2002)
𝑏𝑏 𝑟𝑟
Excess Green Index EXG = (2𝑔𝑔) − 𝑟𝑟 − 𝑏𝑏 YANG et al. (2015)
Excess Red Index EXR = (1,4𝑟𝑟) − 𝑔𝑔 BENDIG et al. (2015)
Excess Green Minus Red Index EXGR = 𝐸𝐸𝐸𝐸𝐸𝐸 − 𝐸𝐸𝐸𝐸𝐸𝐸 GITELSON et al. (2002)
(2𝑔𝑔 − 𝑟𝑟 − 𝑏𝑏)
Green Leaf Index GLI = GITELSON et al. (2002)
(2𝑔𝑔 + 𝑟𝑟 + 𝑏𝑏)
(𝑔𝑔2 − 𝑟𝑟 2 )
Modified Green Red Vegetation Index MGRVI = BENDIG et al. (2015)
(𝑔𝑔2 + 𝑟𝑟 2 )
Modified Photochemical Reflectance (𝑔𝑔 − 𝑟𝑟)
MPRI = BARBOSA et al. (2019)
Index (𝑔𝑔 + 𝑟𝑟)
(𝑔𝑔 − 𝑟𝑟)
Normalized Difference Index NDI = 128 � + 1� MEYER & CAMARGO NETO (2008)
(𝑔𝑔 + 𝑟𝑟)
(𝑔𝑔 − 𝑏𝑏)
Normalized Green-Blue Difference Index NGBDI = BENDIG et al. (2015)
(𝑔𝑔 + 𝑏𝑏)
(𝑔𝑔2 − 𝑟𝑟𝑟𝑟)
Red Green Blue Vegetation Index RGBVI = BENDIG et al. (2015)
(𝑔𝑔2 + 𝑟𝑟𝑟𝑟)
𝑟𝑟
Red Green Index RGI = BENDIG et al. (2015)
𝑔𝑔
= 0,5 [(r-b) – (r-g)]– [(r-g) – (r-
Triangular Greenness Index TGI BENDIG et al. (2015)
b)]
(𝑔𝑔 − 𝑟𝑟)
Visible Atmospherically Resistant Index VARI = GITELSON et al. (2002)
(𝑔𝑔 + 𝑟𝑟 − 𝑏𝑏)
𝑔𝑔
Vegetative Index VEG = 0,667 0,333 HAGUE et al. (2006)
(𝑟𝑟 𝑏𝑏 )
(𝑔𝑔 − 𝑏𝑏)
Woebbecke Index WI = WOEBBECKE et al. (1995)
(𝑟𝑟 − 𝑔𝑔)

r, g e b: normalized RGB bands (equations 1 to 3).

Ciência Rural, v.51, n.8, 2021.

 Evaluation of the nutritional status of corn by vegetation indices via aerial images. 5

orthomosaic RGB bands, employing the QGIS raster analyze the data. This step was done using the QGIS
calculator (QGIS, 2016). First, the extraction of v 2.18 “divider polygon” (QGIS, 2016).
the RGB bands was done from the orthomosaic Pearson’s correlation analysis was done of
and normalized between 0 and 1 (equations 1 to 3) the mean values of the vegetation indices with the N
(MARCIAL-PABLO et al., 2019). and K2O levels present in the soil, as well as the N
and K+ present in the leaves. The analyses of variance
(1) and regression were done to assess the response of
the vegetation indices, N and K+ concentration in the
leaves and grain yield after the treatment was applied.
The statistical analysis using the ExpDes.pt package
from R (FERREIRA et al., 2014) was conducted. For
(2)
those variables which revealed significant interaction
between the N and K2O levels, the response surfaces
were produced together with the supplement of Excel
Real Statistics Resource Pack (ZAIONTZ, 2020)
(3) and the Surfer® software. In figure 2A the flowchart
shows the steps of the process.
Where r, g and b: normalized RGB bands; First, the vegetation indices were adjusted
Rmax = Gmax = Bmax = 255 in 8-bit color images to the polynomial regression models, and the degree
per band. to which this adjustment occurred was evaluated by
The values of the vegetation indices were the coefficient of determination (R2), standard error of
extracted using the QGIS v 2.18 zonal statistical the regression (S) (equation 4), square root of the mean
plugin (QGIS, 2016). For each plot, the zonal statistics square error (RMSE) (equation 5) and the normalized
plugin gives a series of statistical attributes including percentage of RMSE (nRMSE) (equation 6). The S
the maximum, minimum, average, and standard represents the average distance of the recorded values
deviation values. To achieve this, a vector file which with respect to the regression line. The RMSE indicates
contained the useful area of the experimental plots the magnitude of the recorded values versus the values
was used. To create this vector file, the useful area determined by the models, while the nRMSE refers to
of the experimental plot was distinguished into two the normalized measure of the RMSE, which enabled
parts, producing six polygons (subplots), each having a comparison of the performances of the different
an area of 4.0 m2, which were utilized to statistically regression models (LI et al., 2014).

Figure 2 - Flowchart containing the image processing steps.

Ciência Rural, v.51, n.8, 2021.

6 Junior et al.

Other studies reported no response in

(4) the corn from the perspective of grain yield to the
potassium applied to the soil. In fact, BASTOS et
al. (2005) observed no productive response in corn,
(5) hybrid BRS-3123, in their assessment after five levels
of N (0, 50, 100, 150 and 200 kg ha-1) and five levels
(6) of K2O (0, 30, 60, 90 and 120 kg ha-1) were applied
to the soil categorized as Oxisol Yellow-Alic, and
Where n represents the number of sandy / medium in texture. In their work, MELO
observations, Yi refers to the observed values of y, et al. (2011) found that potassium fertilization on
Yi’ are the values assessed by the regression models, maize gave positive effects which were confirmed
Xi includes the observed values of x, Ymax is the in sandy soils, as well as in soils having a K+ level
maximum observed y value, and Ymin is the minimum below 2.3 mmolc dm-3, up to a depth of 0-0.2 m.
observed y value. In such conditions, a dosage of up to 60 kg ha-1 of
K2O induced the best response. In the experimental
RESULTS AND DISCUSSION area, the soil contains a K+ level of 2.0 mmolc dm-3,
in the 0-0.2 m layer, almost within the limit of the
The N and K levels in leaves and grain yield potassium response.
From the analysis of variance, it was Regarding the leaf N content, 28.6 g
evident that the N content in the leaves (NF), the K+ kg-1 was the maximum value recorded after 175.0
level in the leaves (KF) and grain yield (PGS) showed a kg ha-1 of N was applied to the soil; the maximum
response solely to the soil N levels (P < 0.001) (Table 2). KF content (29.5 g kg-1) was reached after 122.3 kg
The results from a few studies revealed that corn shows ha-1 of N was added to the soil. In fact, some studies
greater response when nitrogen is applied to the soil, noted that with the rise in the soil N levels a quadratic
than to when the potassium is added (CARDOSO et increment was seen in the total nitrogen content of
al., 2007; MELO et al., 2011). When 165.0 kg ha-1 of the leaves. Another study by MELO et al. (2011)
N was applied to the soil the maximum PGS (8,536.8 reported maximum NF values in corn, (of the simple
kg ha-1) was obtained (Figure 3). In a study to assess the hybrid BRS 1001 variety), of 28.0 g kg-1, after 175.0
N levels (0, 50, 100, 150 and 200 kg ha-1) and corn kg ha-1 of N was applied in relation to 7.5 plants m-2,
seeding densities (2.5; 5.0; 7.5 and 10.0 plants m-2), concurring with the findings of the current study.
using hybrid BR3060, CARDOSO et al., (2007) When examining the potassium in the
reported maximum grain yield of 8893.0 kg ha-1, leaves, it became evident that the quadratic response
after applying 160.6 kg ha-1 of N related to a sowing was induced only due by the soil N levels and not
density of 7.45 plants m-2, a value which almost by the K2O levels, as mentioned in the literature
corresponds to the maximum value observed in the (PETTER et al., 2016). In fact, PETTER et al. (2016)
current work, with sowing density of 10 plants m-2. in their study on dystrophic Yellow Latosol, having

Table 2 - Analysis of variance (MS) for the nitrogen content of the leaves (NL), potassium in the leaves (K+) (KL) and grain yield (GY)
as a function of the levels of N and K2O in the soil.

SV DF -------------NL--------------- ------------KL-------------- ---------------GY-----------------

* * *
Blocks 2 3.7950 37.420 1810519
*** *** ***
Nitrogen (N) 3 15.4022 51.866 19439535
Potassium (K2O) 3 0.7095 ns 15.726 ns 115619 ns
N versus K2O 9 2.0585 ns 5.943 ns 651722 ns
Residue 30 0.9200 7.038 440954
CV (%) 3.47 9.26 11.47

SV: source of variation; N: nitrogen levels in the soil; K2O: potassium levels in the soil; DF: degrees of freedom; CV (%): coefficient of
variation (%); NF: N content in leaves; KF: K+ content in the leaves; GY: grain yield at 13% humidity. Significance levels by the F test:
ns p > 0.1; ° p < 0.1; * p < 0.05; **p < 0.01; *** p < 0.001.

Ciência Rural, v.51, n.8, 2021.

 Evaluation of the nutritional status of corn by vegetation indices via aerial images. 7

Figure 3 - Regression equations for leaf N content, grain yield and leaf K+ content as a
function of soil N levels. Magalhães de Almeida, MA, 2019.

a sandy-loam texture, noted a significant linear in relation to N (Figure 4A). The other indices also
rise in the K+ level in the corn leaves, in response revealed a significant correlation; however, with the
to increasing the K2O levels added to the soil. A r values below 0.4, such as MGRVI, MPRI, GLI,
maximum K2O level of 120.0 kg ha-1 induced the RGI, EXG, CIVE and CI (Figure 4A). As for the K+
leaves to accumulate 25.4 g kg-1 of K+. However, content in the leaves, the RGI index alone showed
this K+ level rise in the leaves had no effect on significant correlation (p < 0.05). The r = 0.222 was
the relative total chlorophyll content, revealing the regarded as low, according to the classification of
absence of any direct correlation between the K+ HOPKINS (2000), which disqualifies it as a good
concentration in the leaves and chlorophyll synthesis. indicator for the detection of the K+ level in the maize
However, in the present study, this trend was not leaves (Figure 4B).
observed, likely because the soil K+ concentration These vegetation indices are understood to
in the experimental area prevented the expression of offer promise, even if the use of the RGB bands alone
the K2O levels applied, as emphasized earlier. was sufficient to distinguish between the spectral
responses of the corn canopy, depending upon the
Pearson correlation between the N, K+ content in the N doses added to the soil. Variations in detecting the
leaves and vegetation indices spectral response of the corn canopy via vegetation
The correlation found between the N indices with regards to the levels of soil fertility and
and K+ leaf contents and the N and K2O soil levels N concentration in the leaves were also noted in the
were r = 0.684 (p < 0.001) and r = 0.284 (p < 0.05), research performed by LI et al. (2014), CILIA et al.
respectively, establishing the higher N response in (2014) and VERGARA-DÍAZ et al. (2016).
terms of the yield performance of corn, as discussed In fact, VERGARA-DIAZ et al. (2016)
prior (Table 2). Regarding the N concentration in derived different vegetation indices using the digital
the leaves, a significant correlation was seen for RGB images at the levels of the leaf and canopy and
ten indices, particularly on the EXR (r = -0.479; p assessed as inexpensive tools in corn fertilization
< 0.001), NDI (r = 0.454; p < 0.001) and VARI (r = management. The effectiveness of the RGB indices
0.412; p < 0.001), which are shown to be the most was compared with that of the other indices that
promising in detecting the nutritional status of corn employ bands in the near infrared region, like the

Ciência Rural, v.51, n.8, 2021.

8 Junior et al.

Figure 4 - Pearson’s correlation between vegetation indices and the content of N and K+ in the leaves in response to the levels of N
and K2O applied in the soil. A: N content in leaves; B: K+ content in the leaves. Significance levels by the t test: ns p > 0.1;
° p < 0.1; * p< 0.05; ** p< 0.01; *** p < 0.001.

NDVI (Normalized Difference Vegetation Index), different levels of soil N (0, 80, 60 and 320 kg ha-1)
and the leaf chlorophyll concentration (LCC - employing RGB vegetation indices. They determined
Leaf Chlorophyll Content) during flowering. The that not only did the VARI index best discriminate the
conclusion drawn by the authors was that the grain nitrogen added, at all the times of the evaluation being
productivity and N levels in the leaves at the evaluated investigated, it also revealed the highest correlation
levels of fertilization were emphatically anticipated with the readings of the chlorophyll and dry mass.
by the majority of the RGB indices (with R2 ± 0.7), Later, ISLA et al. (2011) indicated that the GNDVI
closely corresponding to the NDVI and LCC. and GRVI indices showed much promise as well, in
It was CILIA et al. (2014) who employed the detection of the nutritional status of corn with
the vegetation indices of the hyperspectral remote respect to N, particularly during the developmental
sensing images to assess the techniques of mapping stages of V6-V8. It was GHOLIZADEH et al. (2011)
the nitrogen levels in the corn crop. Based on the who identified a high degree of correlation between
Nitrogen Nutrition Index (NNI) the nitrogen status the EXR index and N level in the rice leaves induced
was determined, interpreted as the ratio between the by the soil N levels (0, 85 and 170 kg ha-1).
N content of the leaves and the minimum N content It was also noteworthy that the Pearson’s
necessary for the maximum dry biomass yield. correlation values of 0.3 to 0.6 were commonly
The best performances were seen by the MCARI identified in works done to assess the nutritional
/ MTVI2 index (Modified Chlorophyll Absorption status of agricultural crops by the use of RGB
Proportion Index / Modified Triangular Vegetation vegetation indices (LI et al., 2014; CILIA et al., 2014;
Index 2) in evaluating the N content in the leaves VERGARA-DÍAZ et al., 2016; RASMUSSEN et
(R2 = 0.59) and MTVI2 in determining the dry al., 2016).
biomass (R2 = 0.80). The NNI map concurred with
the estimated NNI using field data, employing the Vegetation indices in response to soil N and K2O levels
traditional destructive measurements (R2 = 0.70), From the analysis of variance, the interaction
to confirm the potential of using the remotely detected between the N and K2O showed significance for the
indices in the assessment of the nutritional status of vegetation indices EXR (p < 0.05) and NDI (P <
corn related to N levels. 0.05), while the VARI (p < 0.001) responded to the
In their study, ABRAHÃO et al. (2009) N application alone (Table 3). It is significant that
estimated the nutritional status of Tanzania grass at the EXR and NDI indices could detect the spectral

Ciência Rural, v.51, n.8, 2021.

 Evaluation of the nutritional status of corn by vegetation indices via aerial images. 9

Table 3 - Analysis of variance (MS) for the vegetation indices in response to the levels of N and K2O in the soil. Magalhães de Almeida,
MA, 2019.

SV DF -------------EXR------------ ---------------NDI-------------- --------------VARI---------------

*** *** ***
Blocks 2 0.00642 277.850 0.01515
*** *** ***
Nitrogen (N) 3 0.00328 76.426 0.00332
Potassium (K2O) 3 0.00004 ns 6.714 ns 0.00078 ns
* *
N versus K2O 9 0.00070 16.464 0.00068 ns
Residue 30 0.00024 5.409 0.00050
CV (%) 12.03 1.72 23.20

SV: source of variation; N: nitrogen levels in the soil; K2O: potassium levels in the soil; DF: degrees of freedom; CV (%): coefficient of
variation (%); EXR: Excess Green Index; NDI: Normalized Difference Index; VARI: Visible Atmospherically Resistant Index.
Significance levels by the F test: ns p>0.1; ° p<0.1; * p<0.05; ** p<0.01; *** p<0.001.

response of the interaction between the soil N and When the EXR value is 0.146 it means that the soil
K2O concentrations; however, this interaction was has both N and K2O in low levels (0 to 20 kg ha-1);
absent with respect to the agronomic response to the however, when the EXR is 0.118 it suggests that the
N and K+ levels in the leaves and grain productivity soil N levels (160 to 180 kg ha-1) and K2O levels (120
(Table 2). to 140 kg ha-1) are high. With respect to the NDI, a
For the EXR index, the response surface value of 134.0 indicates that the soil has low N and
was adjusted to a 1st degree polynomial model K2O levels (0 to 20 kg ha-1). When the NDI value is
(y = a + bx + by + dxy), with R2 = 0.656, with 138.8 it implies that the soil has high N (160 to 180 kg
significance at 5% by the F test, with the model ha-1) and K2O levels (120 to 140 kg ha-1).
coefficients showing significance at 0.1% (b) and 10% High levels of soil N and K2O are indicated
(c and d), by t test, and the standard error of estimate when the EXR value is 0.117 and the NDI is 138.8.
equal to 0.0104. The EXR values are lowered as the This is linked to a high average grain productivity of
rise in the soil N and K2O levels reach up to 100 kg around 8,520.4 kg ha-1, while the values of EXR and
ha-1 of N, following which the decrease in the EXR NDI, 0.146 and 134.0, respectively are linked to low
begins to take place by a drop in the soil K2O levels. average grain production (5,430.2 kg ha-1). This has
The NDI showed a similar trend which was also been attributed to the low N and K2O availability in
adjusted to the 1st degree polynomial model, just the soil (Figure 3). Therefore, for the farmer, the use
as it was adjusted for the EXR (Figures 5A and 5B). of remote detection of the nutritional status of corn
A nutritional evaluation work done by VERGARA- through the EXR and NDI indices is very useful in
DIAZ et al. (2016) on corn in relation to N (0, 10, helping him to decide whether to increase or decrease
20, 80 and 160 kg ha-1) using RGB images, found the application of the nitrogen and / or potassium
that the indices showing promise in identifying the N fertilizers, to optimize the grain yield.
showed sensitivity to the variations in the N content The VARI index adjusted to a polynomial
of leaves up to the 80 kg ha-1 level of N; however, model of the 1st degree, as a response to the soil N
none of them was significantly related to the 160 levels, shows higher quality of the indicators R2,
kg ha-1 level of N. RMSE, nRMSE and S when compared with the
The highest EXR (0.146) value was seen other models (Figure 6). High VARI values of 0.1283
with the combined lowest doses of N and K2O (0 N suggest high N levels in the soil (180 kg ha-1), which
- 0 K2O); the lowest value (0.104) value was noted induces high grain productivity on average (8,520.4
with 180 kg ha-1 level of N and 0 kg ha-1 level of K2O. kg ha-1) (Figure 3). However, the low VARI values
From the isoquants, the ranges of the EXR and NDI (0.097) reveal lowered N levels in the soil, which
values associated with the different levels of N and are unfavorable to achieving satisfactory grain yield
K2O present in the soil can be identified (Figures (5,430.2 kg ha-1) (Figure 3).
5C and 5D), which facilitates the detection of the It was CILIA et al. (2014) and VERGARA-
nutritional status of corn based on these indices. DÍAZ et al. (2016) who obtained the linear relationship

Ciência Rural, v.51, n.8, 2021.

 Evaluation of the nutritional status of corn by vegetation indices via aerial images. 10

Figure 5 - Response surfaces for the EXR (A, C) and NDI (B, D) indices as a function of the levels of N and K2O in the soil. Magalhães
de Almeida, MA, 2019.

between the N concentration in the corn leaves and CONCLUSION

vegetation indices. This trend was also noted for the
N level in the leaves and the grain produced (Figure Significant correlation was observed for
3). In fact, VIÑA et al. (2004) in their assessment of the EXR, NDI and VARI indices with the leaf N
the phenological development of corn with the help content, which endorses them as encouraging in the
of RGB images, came to understand that the VARI identification of the nutritional status of corn with
index is highly sensitive to the response to the leaf respect to N.
chlorophyll content. From these authors it is evident The nutritional status of corn could not be
that this index may be indicative of an early stress detected with regards to potassium, by even one of
phase in the crop because one of the symptoms the indices assessed. The EXR and NDI indices were
suggestive of stress with respect to N is the drop in able to capture the interaction between the N and K2O
the leaf chlorophyll content (SRIDEVY et al., 2018). levels in the soil.

Ciência Rural, v.51, n.8, 2021.

 Evaluation of the nutritional status of corn by vegetation indices via aerial images. 11

Figure 6 - Regression to the VARI index in response to soil N levels. Magalhães de Almeida,
MA, 2019.

ACKNOWLEDGMENTS content/uploads/2019/05/Vol17No2_Barbosa.pdf#abstract-6898>.
Accessed: May, 08, 2020. doi: 10.15159/AR.19.119.
The authors express their gratitude to Weisul Agrícola
for providing the experimental area and support during the BASTOS, E. A. et al. Produtividade de grãos de milho sob
installation and performance of the experiment. diferentes doses de nitrogênio e potássio em solos de cerrado
do sudoeste piauiense. Teresina: Embrapa Meio-Norte, 2005. 17p.
(Boletim de Pesquisa e Desenvolvimento, 59). Available from:
DECLARATION OF CONFLICT OF <https://www.infoteca.cnptia.embrapa.br/infoteca/bitstream/
INTEREST doc/68844/1/BOLP59.pdf>. Accessed: May, 07, 2020.

The authors have no conflict of interests to declare. BENDIG, J. et al. Combining UAV-based plant height from crop
Sponsors did not play any part in the design of the study. They surface models, visible, and near infrared vegetation indices for
neither had any role in the data collection, its analysis or biomass monitoring in barley. International Journal of Applied
interpretation nor in the manuscript writing and the decision to Earth Observation and Geoinformation, v.39, p.79-87, 2015.
publish the findings. Available from: <https://www.sciencedirect.com/science/article/
abs/pii/S0303243415000446>. Accessed: May, 12, 2020. doi:
10.1016/j.jag.2015.02.012.
AUTHORS’ CONTRIBUTIONS
CARDOSO, M. J. et al. Níveis de nitrogênio e densidade de
All the authors have made equal contributions plantas de milho em sistema plantio direto nos municípios de
towards the design and writing of the manuscript. All the authors Baixa Grande do Ribeiro, PI e São Raimundo das Mangabeiras,
have critically reviewed the manuscript and given their approval MA. Teresina: Embrapa Meio-Norte, 2007. 15p. (Boletim de
for the final version. Pesquisa e Desenvolvimento, 69). Available from: <https://
www.infoteca.cnptia.embrapa.br/infoteca/handle/doc/67702>.
Accessed: May, 07, 2020.
REFERENCES
CILIA, C. et al. Nitrogen status assessment for variable rate
ABRAHÃO, S. A. et al. Vegetation spectral indices to discriminate fertilization in maize through hyperspectral imagery. Remote
nitrogen rates in tanzania grass. Revista Brasileira de Zootecnia, Sensing, v.6, p.6549-6565, 2014. Available from: <https://www.
v.38, n.9, p.1637-1644, 2009. Available from: <https://www.scielo. mdpi.com/2072-4292/6/7/6549>. Accessed: Mar. 28, 2020. doi:
br/pdf/rbz/v38n9/01.pdf>. Accessed: May, 20, 2020. 10.3390/rs6076549.

BARBOSA, B. D. S. et al. RGB vegetation indices applied to grass CORREIA FILHO, F. L. et al. Projeto cadastro de fontes de
monitoring: a qualitative analysis. Agronomy Research, v.17, n.2, abastecimento por água subterrânea, estado do Maranhão:
p.349–357, 2019. Available from: <https://agronomy.emu.ee/wp- relatório diagnóstico do município de Magalhães de Almeida.

Ciência Rural, v.51, n.8, 2021.

 Evaluation of the nutritional status of corn by vegetation indices via aerial images. 12

Teresina: CPRM - Serviço Geológico do Brasil, 2011. 31p. from: <https://www.tandfonline.com/doi/abs/10.1080/01431161.2

Available from: <https://http://rigeo.cprm.gov.br/xmlui/bitstream/ 018.1528017?journalCode=tres20>. Accessed: Apr. 13, 2020. doi:
handle/doc/15524/rel-magalhaes_almeida.pdf?sequence=3>. 10.1080/01431161.2018.1528017.
Accessed: Mar. 10, 2020.
MELO, F. B. et al. Nitrogen fertilization, plant density and
FERREIRA, E. et al. ExpDes: An R Package for ANOVA and maize yield cropped under no-tillage system. Revista Ciência
Experimental Designs. Applied Mathematics, v.5, n.19, p.2952- Agronômica, Fortaleza, v.42, n.1, p.27-31, 2011. Available
2958, 2014. Available from: <https://https://www.scirp.org/ from: <http://ccarevista.ufc.br/seer/index.php/ccarevista/article/
journal/paperinformation.aspx?paperid=51204>. Accessed: Mar. view/951>. Accessed: May, 07, 2020.
10, 2020. doi: 10.4236/am.2014.519280.
MEYER, G. E. et al. Verification of color vegetation indices
GHOLIZADEH, A. et al. Temporal variability of SPAD chlorophyll for automated crop imaging applications. Computers and
meter readings and its relationship to total nitrogen in leaves Electronics in Agriculture, v.63, p.282-293, 2008. Available
within a Malaysian paddy field. Australian Journal of Basic from: <https://www.sciencedirect.com/science/article/
Applied Science, v.5, n.5, p.236-245, 2011. Available from: pii/S0168169908001063>. Accessed: May, 08, 2020. doi:
<http://www.ajbasweb.com/old/ajbas/2011/236-245.pdf>. 10.1016/j.compag.2008.03.009.
Accessed: Feb. 20, 2020.
PETTER, F. A. et al. Doses and time of potassium application
GITELSON, A. A. et al. Novel algorithms for remote estimation on corn agronomic performance cultivated in a ‘Cerrado’ area at
of vegetation fraction. Remote Sensing of Environment, v.80, Piaui State, Brazil. Comunicata Scientiae, v.7, n.3, p.372-382,
n.1, p.76–87, 2002. Available from: <https://www.sciencedirect. 2016. Available from: <https://www.comunicatascientiae.com.
com/science/article/pii/S0034425701002899>. Accessed: Feb. 20, br›article›download>. Accessed: May, 16, 2020. doi: 10.14295/
2020. doi: 10.1016/S0034-4257(01)00289-9. CS.v7i3.1218.

HAGUE, T. et al. Automated crop and weed monitoring in QGIS Development Team. 2016. QGIS 2.18. Geographic
widely spaced cereals. Precision Agriculture, v.7, p.21–32, Information System: User Guide. Open Source Geospatial
2006. Available from: <https://link.springer.com/article/10.1007/ Foundation Project. Available from: <https://docs.qgis.org/2.18/
s11119-005-6787-1>. Accessed: Apr. 13, 2020. doi: 10.1007/ pdf/pt_BR>. Accessed: Jan. 11, 2016.
s11119-005-6787-1.
RASMUSSEN, J. et al. Are vegetation indices derived from
HOPKINS, W. G. Correlation coefficient: a new view of statistics. consumer-grade cameras mounted on UAVs sufficiently reliable for
2000. Available from: <http://www.sportsci.org/resource/stats/ assessing experimental plots? European Journal of Agronomy,
correl.html>. Accessed: Mar. 12, 2020. v.74, p.75–92, 2016. Available from: <https://www.sciencedirect.
com/science/article/pii/S1161030115300733>. Accessed: May, 08,
ISLA, R. et al. Utilización de imágenes aéreas multiespectrales 2020. doi: 10.1016/j.eja.2015.11.026.
para evaluar la disponibilidad de nitrógeno en maíz. In:
RECONDO, C.; PENDÁS, E. Eds. Teledetección, Bosques SANTOS, H. G. et al. Sistema Brasileiro de Classificação
y Cambio Climático. Asociación Española de Teledetección: de Solos. 5. ed. rev. e ampl. Brasília, DF: Embrapa, 2018.
Mieres, España, 2011, p.9–12. Available from: <https://digital.csic. 356p. Available from: <https://www.embrapa.br/solos/busca-
es/bitstream/10261/42274/1/ISLAR_CongresoAET_%202011. de-publicacoes/-/publicacao/1094003/sistema-brasileiro-de-
pdf>. Accessed: May, 20, 2020. classificacao-de-solos>. Accessed: 11 Feb., 2020.

LAGRANGE et al. Large-scale feature selection with Gaussian SILVA, F. C. Manual de análises químicas de solos, plantas e
mixture models for the classification of high dimensional remote fertilizantes. Brasília, DF: Embrapa Informação Tecnológica;
sensing images. IEEE Transactions on Computational Imaging, Rio de Janeiro: Embrapa Solos, 2009. 627p. Available
IEEE, v.3, n.2, p.230-242, 2017. Available from: <https://hal. from: <https://www.infoteca.cnptia.embrapa.br/bitstream/
archives-ouvertes.fr/hal-01382500v4>. Accessed: Jan. 20, 2020. doc/990374/1/ManualdeMtodosdeAnilisedeSolo.pdf>. Accessed:
doi: 1109/TCI.2017.2666551. Jan. 11, 2020.

LI, F. et al. Improving estimation of summer maize nitrogen SRIDEVY, S. et al. Nitrogen and potassium deficiency
status with red edge-based spectral vegetation indices. Field identification in maize by image mining, spectral and true color
Crops Research, v.157, p.111-123, 2014. Available from: <https:// response. Indian Journal of Plant Physiology, v.23, p.91-99,
www.sciencedirect.com/science/article/pii/S0378429013004322>. 2018. Available from: <https://link.springer.com/article/10.1007/
Accessed: Apr. 13, 2020. doi: 10.1016/j.fcr.2013.12.018. s40502-018-0359-7>. Accessed: May, 13, 2020. doi: 10.1007/
s40502-018-0359-7.
MANDAL, U. K. Spectral color indices-based geospatial
modeling of soil organic matter in Chitwan District, Nepal. TORRES-SANCHEZ, J. et al. An automatic object-based method
In: REMOTE SENSING AND SPATIAL INFORMATION for optimal thresholding in UAV images: Application for
SCIENCES, 23., Prague, 2016. Proceedings. Prague: ISPRS, vegetation detection in herbaceous crops. Computers and
2016, p.43-48. Available from: <https://ui.adsabs.harvard.edu/ Electronics in Agriculture, v.114, p.43-52, 2015. Available
abs/2016ISPAr49B2...43M/abstract>. Accessed: May, 12, 2020. from: <https://www.sciencedirect.com/science/article/pii/
doi: 10.5194/isprs-archives-XLI-B2-43-2016. S0168169915001052>. Accessed: Apr. 13, 2020. doi: 10.1016/j.
compag.2015.03.019.
MARCIAL-PABLO, M. J. et al. Estimation of vegetation fraction
using RGB and multispectral images from UAV. International VERGARA-DÍAZ, O. et al. A novel remote sensing approach for
Journal of Remote Sensing, v.40, n.2, p.420–438, 2019. Available prediction of maize yield under different conditions of nitrogen

Ciência Rural, v.51, n.8, 2021.

 Evaluation of the nutritional status of corn by vegetation indices via aerial images. 13

fertilization. Frontiers in Plant Science, v.7, p.666-000, 2016. conditions. Transactions of the ASAE, 38: 259-269, 1995.
Available from: <https://www.ncbi.nlm.nih.gov/pmc/articles/ Available from: <https://elibrary.asabe.org/abstract.
PMC4870241/>. Accessed: May, 08, 2020. doi: 10.3389/ asp?aid=27838>. Accessed: Feb. 25, 2020.
fpls.2016.00666.
YANG, W. et al. Greenness identification based on HSV decision
VIÑA, A. et al. Monitoring maize (Zea mays L.) phenology with tree. Information Processing in Agriculture, v.2, n.3-4,
remote sensing. Agronomy Journal, v.96, p.1139-1147, 2004. p.149–160, 2015. Available from: <https://www.sciencedirect.
Available from: <https://digitalcommons.unl.edu/cgi/viewcontent. com/science/article/pii/S2214317315000347>. Accessed: May,
cgi?article=1266&context=natrespapers>. Accessed: May, 12, 2020. 31, 2020. doi: 10.1016/j.inpa.2015.07.003.

WOEBBECKE, D. M. et al. Color indices for weed ZAIONTZ, C. Real Statistics Using Excel. Available from:
identification under various soil, residue, and lighting <www.real-statistics.com>. Accessed: Jan. 10, 2020.

Ciência Rural, v.51, n.8, 2021.