Data processing to remove outliers and inliers: A systematic literature study
Revista Brasileira de Engenharia Agrícola e Ambiental,
Год журнала:
2024,
Номер
28(9)
Опубликована: Янв. 1, 2024
ABSTRACT
Outliers
and
inliers
often
arise
during
sample
data
acquisition.
While
outliers
represent
anomalous
observations,
are
erroneous
points
within
the
main
body
of
dataset.
It
was
aimed
to
conduct
a
systematic
literature
study
(SLS)
survey
methods
software
employed
for
outlier
inlier
removal,
particularly
exploratory
analysis.
The
conducted
in
three
phases:
(i)
mapping
(SLM),
(ii)
snowballing
(SB),
(iii)
SLR.
Initially,
772
scientific
studies
were
identified,
subsequently
narrowed
down
86
after
applying
selection
criteria.
Backward
(BSB)
forward
(FSB)
further
yielded
16
studies,
resulting
final
pool
102
identified
removal
techniques
(Chebyshev’s
inequality,
boxplot,
principal
component
analysis),
one
technique
(local
Moran’s
index),
thirteen
commonly
used
software.
Язык: Английский
Assessing the Impact of Overhead Agrivoltaic Systems on GNSS Signal Performance for Precision Agriculture
Smart Agricultural Technology,
Год журнала:
2024,
Номер
unknown, С. 100664 - 100664
Опубликована: Ноя. 1, 2024
Язык: Английский
Use of supervised and unsupervised approaches to make zonal application maps for variable-rate application of crop growth regulators in commercial cotton fields
Journal of Cotton Research,
Год журнала:
2025,
Номер
8(1)
Опубликована: Янв. 6, 2025
Abstract
Background
Zonal
application
maps
are
designed
to
represent
field
variability
using
key
variables
that
can
be
translated
into
tailored
management
practices.
For
cotton,
zonal
for
crop
growth
regulator
(CGR)
applications
under
variable-rate
(VR)
strategies
commonly
based
exclusively
on
vegetation
indices
(VIs)
variability.
However,
VIs
often
saturate
in
dense
areas,
limiting
their
effectiveness
distinguishing
growth.
This
study
aimed
compare
unsupervised
framework
(UF)
and
supervised
(SUF)
approaches
generating
CGR
VR
conditions.
During
2022–2023
agricultural
seasons,
an
UF
was
employed
generate
locally
collected
data
plant
height
of
satellite
imagery,
soil
texture,
phenology
data.
Subsequently,
a
SUF
(based
historical
between
2020–2021
seasons)
developed
predict
remote
sensing
data,
aiming
replicate
same
but
without
relying
direct
measurements
height.
Both
were
tested
three
fields
two
different
dates
per
field.
Results
The
predictive
model
performed
well,
as
indicated
by
the
metrics.
when
comparing
specific
field-date
combinations,
predicted
exhibited
lower
compared
with
measurements.
led
variable
compatibility
maps,
which
utilized
predictions,
real
Fields
characterized
much
pronounced
texture
yielded
highest
produced
both
approaches.
predominantly
due
greater
consistency
estimating
development
patterns
within
these
heterogeneous
environments.
While
approach
facilitate
product
savings
during
operation,
other
factors
must
considered.
These
include
availability
specialized
machinery
required
this
type
applications,
well
inherent
operational
costs
associated
applying
single
differs
from
typical
uniform
rate
integrate
multiple
inputs.
Conclusion
Predictive
modeling
shows
promise
assisting
creation
applications.
degree
agreement
actual
found
should
evaluated
field-by-field
basis.
approach,
is
heigh
prediction,
demonstrated
potential
supporting
aligns
itself
observed
may
vary,
necessitating
evaluation.
Язык: Английский
Numerical simulation and optimization design of a novel longitudinal-flow online fertilizer mixing device
Computers and Electronics in Agriculture,
Год журнала:
2025,
Номер
237, С. 110546 - 110546
Опубликована: Май 30, 2025
Язык: Английский
A new method for satellite-based remote sensing analysis of plant-specific biomass yield patterns for precision farming applications
Precision Agriculture,
Год журнала:
2024,
Номер
25(6), С. 2801 - 2830
Опубликована: Апрель 28, 2024
Abstract
This
study
describes
a
new
method
for
satellite-based
remote
sensing
analysis
of
plant-specific
biomass
yield
patterns
precision
farming
applications.
The
relative
potential
(rel.
BMP)
serves
as
an
indicator
multiyear
stable
and
homogeneous
zones.
rel.
BMP
is
derived
from
satellite
data
corresponding
to
specific
growth
stages
the
normalized
difference
vegetation
index
(NDVI)
analyze
crop-specific
patterns.
development
this
methodology
based
on
arable
fields
two
research
farms;
validation
was
conducted
commercial
farms
in
southern
Germany.
Close
relationships
(up
r
>
0.9)
were
found
between
different
crop
types
years,
indicating
fields.
showed
moderate
correlations
=
0.64)
with
yields
determined
by
combine
harvester,
strong
red
edge
inflection
point
(REIP)
0.88,
tractor-mounted
sensor
system)
sampling
0.57).
investigated
relationship
key
soil
parameters.
There
consistently
correlation
organic
carbon
(SOC)
total
nitrogen
(TN)
contents
(r
0.62
0.73),
demonstrating
that
effectively
reflects
impact
these
properties
yield.
approach
well
suited
deriving
zones,
extensive
application
agriculture.
Язык: Английский
VegIndex: rotina computacional de código-fonte aberto do Google Earth Engine para análise espaço-temporal de índice de vegetação
Caderno Pedagógico,
Год журнала:
2024,
Номер
21(7), С. e5995 - e5995
Опубликована: Июль 23, 2024
O
Google
Earth
Engine®
(GEE)
é
uma
plataforma
de
processamento
alta
performance,
análise
e
visualização
dados
geoespaciais
por
meio
computação
em
nuvem.
Apesar
do
rápido
crescimento
na
quantidade
aplicações
desenvolvidas
no
GEE
nos
últimos
anos,
relacionados
à
diversos
temas
da
agricultura
precisão
(AP),
ainda
há
necessidade
desenvolver
mais
específicas
ou
personalizadas.
Nesse
contexto,
instigado
pelo
aumento
desenvolvimento
adoção
ferramentas
digitais
agricultura,
bem
como
pela
tendência
nuvem,
presente
trabalho
objetivou-se
rotina
computacional
automatizada,
open
source
escalável
(repositório
código
com
todos
os
scripts
GEE)
para
realizar
análises
espaço-temporal
índice
vegetação
NDVI.
Para
tanto,
foram
utilizados
três
datasets
(ou
coleções)
oriundos
dos
instrumentos
Landsat
8,
9
Sentinel-2,
disponíveis
repositório
GEE.
A
desenvolvida,
denominada
VegIndex,
possui
módulos
processamento,
dados,
estruturados
programação
orientada
a
objetos
(funções),
linguagem
JavaScript
editada
Code
Editor
Esses
não
são
sequenciais,
seja,
podem
ser
executados
independentemente.
supracitada
foi
testada
área
comercial
(122,70
ha)
cultivo
irrigado
algodoeiro.
desenvolvida
permitiu
NDVI,
forma
rápida
intuitiva,
grande
potencial
uso
precisão.
Além
disso,
source,
permite
o
usuário
(pesquisador,
consultor,
técnico,
produtor,
etc)
adaptá-la
diversas
culturas
agrícolas
espécies
florestais.
VegIndex
exige
interferência
configurações
das
funções,
filtros
algoritmos
geoprocessamento
(100%
automatizados).