Land,
Journal Year:
2023,
Volume and Issue:
12(10), P. 1934 - 1934
Published: Oct. 18, 2023
Plastic
in
agriculture
is
frequently
used
to
protect
crops
and
its
use
boosts
output,
enhances
food
quality,
contributes
minimize
water
consumption,
reduces
the
environmental
impacts
of
agricultural
activities.
On
other
hand,
end-of-life
plastic
management
disposal
are
main
issues
related
their
presence
this
kind
environment,
especially
respect
degradation,
if
not
properly
handled
(i.e.,
storage
places
directly
contact
with
ground,
exposure
stocks
meteoric
agents
for
long
periods,
incorrect
or
incomplete
removal).
In
study,
possibility
using
an
situ
near
infrared
(NIR:
1000–1700
nm)
hyperspectral
imaging
detection
architecture
recognition
various
wastes
soils
order
identify
also
assess
degradation
from
a
recovery/recycling
perspective
was
explored.
more
detail,
Partial
Least
Squares—Discriminant
Analysis
(PLS-DA)
classifier
capable
identifying
waste
soil
developed,
implemented,
set
up.
Results
showed
that
imaging,
combination
chemometric
approaches,
allows
utilization
rapid,
non-destructive,
non-invasive
analytical
approach
characterizing
produced
agriculture,
as
well
potential
assessment
lifespan.
Frontiers in Environmental Science,
Journal Year:
2024,
Volume and Issue:
12
Published: Feb. 20, 2024
Detecting
hazardous
substances
in
the
environment
is
crucial
for
protecting
human
wellbeing
and
ecosystems.
As
technology
continues
to
advance,
artificial
intelligence
(AI)
has
emerged
as
a
promising
tool
creating
sensors
that
can
effectively
detect
analyze
these
substances.
The
increasing
advancements
information
have
led
growing
interest
utilizing
this
environmental
pollution
detection.
AI-driven
sensor
systems,
AI
Internet
of
Things
(IoT)
be
efficiently
used
monitoring,
such
those
detecting
air
pollutants,
water
contaminants,
soil
toxins.
With
concerns
about
detrimental
impact
legacy
emerging
on
ecosystems
health,
it
necessary
develop
advanced
monitoring
systems
detect,
analyze,
respond
potential
risks.
Therefore,
review
aims
explore
recent
using
AI,
IOTs
taking
into
account
complexities
predicting
tracking
changes
due
dynamic
nature
environment.
Integrating
machine
learning
(ML)
methods
revolutionize
science,
but
also
poses
challenges.
Important
considerations
include
balancing
model
performance
interpretability,
understanding
ML
requirements,
selecting
appropriate
models,
addressing
related
data
sharing.
Through
examining
issues,
study
seeks
highlight
latest
trends
leveraging
IOT
monitoring.
TrAC Trends in Analytical Chemistry,
Journal Year:
2023,
Volume and Issue:
166, P. 117221 - 117221
Published: Aug. 3, 2023
Numerous
studies
have
attempted
to
detect
microplastic
litter
directly
in
environmental
sediments
via
spectral
imaging
and
powerful
classification
algorithms.
Spectral
is
attractive
largely
due
the
benefits
of
adding
a
spatial
element
data,
relative
measuring
speed,
minimal
sample
processing.
Despite
this
promise,
important
concerns
related
selectivity
must
be
considered
along
with
appropriateness
Here
we
evaluate
performance
near
infrared
hyperspectral
(NIR-HSI)
four
commonly
used
algorithms
on
simple
test
case
which
images
individual
microplastics
known
size
top
sand
were
collected.
The
results
highlight
major
weak
points
NIR-HSI
machine
learning
as
applied
detection
microplastics,
large
proportion
false
positives
negatives
most
situations
studied,
alerts
reader
about
use
methodology.
Bulletin of the Korean Chemical Society,
Journal Year:
2024,
Volume and Issue:
45(5), P. 472 - 481
Published: March 7, 2024
Abstract
The
escalating
concern
regarding
microplastics
(MPs)
in
the
environment
has
recently
accentuated
need
for
comprehensive
analyses
across
various
matrices.
Fourier
Transfrom
Infrared
(FT‐IR)
microscopy
is
widely
used
method
MP
identification,
but
challenges
arise
due
to
presence
of
secondary
materials
on
real
samples,
causing
inaccuracies
spectral
matching.
To
tackle
this
issue,
we
propose
a
solution:
1D‐convolution
neural
network
(1D‐CNN)
machine‐learning
model
classifying
FT‐IR
spectra
into
16
polymer
species.
Using
dataset
5413
spectra,
with
80%
(4330)
training
and
20%
(1083)
external
testing,
our
achieved
98.59%
accuracy
cross‐validation
92.34%
validation.
This
study
underscores
efficacy
machine
learning
discerning
types
among
MPs,
even
samples
tainted
by
materials.
implementation
1D‐CNN
marks
significant
leap
overcoming
conventional
limitations,
providing
robust
tool
accurately
unraveling
MPs
intricacies
environmental
RSC Advances,
Journal Year:
2023,
Volume and Issue:
13(51), P. 36223 - 36241
Published: Jan. 1, 2023
This
review
highlights
the
range
of
spectroscopic
techniques,
methods
and
tools
developed
for
microplastics
separation,
analysis
their
accumulation
in
various
edible
species
implications
on
our
food
chain.
