bioRxiv (Cold Spring Harbor Laboratory),
Год журнала:
2024,
Номер
unknown
Опубликована: Ноя. 11, 2024
Abstract
This
study
investigated
the
use
of
multi-sourced
photographs,
such
as
those
from
social
media,
camera
traps,
and
field
surveys,
for
individual
identification
male
Asian
elephants
to
their
corridor
use,
crop-raiding
behaviour,
occurrence
musth,
ranging
external
body
injuries
in
Thirunelli-Kudrakote
elephant
India’s
Western
Ghats.
A
total
330
images
videos
were
analysed
over
11
years
(January
2014
March
2024),
leading
27
corridor.
Eleven
individuals
observed
across
multiple
years,
demonstrating
fidelity
area.
Ten
engaging
crop-raiding,
with
four
consistently
raiding
crops
within
The
also
assessed
home
range
6
adjacent
landscapes
based
on
identification.
Overall,
shows
that
identifying
different
locations
times
provides
valuable
data
is
a
practical,
cost-effective
method
investigating
wide
questions
concerning
ecology
behaviour
elephants.
findings
underscore
potential
citizen
science
initiatives
enhance
research
elephant-range
countries.
Mammalian Biology,
Год журнала:
2021,
Номер
102(3), С. 983 - 993
Опубликована: Дек. 27, 2021
Abstract
Identifying
individual
animals
is
critical
to
describe
demographic
and
behavioural
patterns,
investigate
the
ecological
evolutionary
underpinnings
of
these
patterns.
The
traditional
non-invasive
method
identification
in
mammals—comparison
photographed
natural
marks—has
been
improved
by
coupling
other
sampling
methods,
such
as
recording
overhead
video,
audio
multimedia
data.
However,
aligning,
linking
syncing
data
streams
are
persistent
challenges.
Here,
we
provide
computational
tools
streamline
integration
multiple
techniques
identify
free-ranging
mammals
when
tracking
their
behaviour
wild.
We
developed
an
open-source
R
package
for
organizing
simplifying
processing
a
posteriori—“
MAMMals
:
Managing
Animal
MultiMedia:
Align,
Link,
Sync”.
contains
functions
(i)
align
link
from
photographs
videos,
recordings
text
sources
(e.g.
GPS
locations)
which
metadata
can
be
accessed;
(ii)
synchronize
extract
useful
videos
with
audios)
containing
photo-identified
individuals.
To
illustrate
how
facilitate
photo-identification
video
situ,
simultaneously
collected
photos
bottlenose
dolphins
using
off-the-shelf
cameras
drones,
then
merged
track
foraging
individuals
groups.
hope
our
simple
encourage
future
work
that
extend
generalize
links
between
platforms
mammals,
thereby
improving
raw
material
needed
generating
new
insights
mammalian
population
ecology.
Wildlife Research,
Год журнала:
2024,
Номер
51(11)
Опубликована: Окт. 21, 2024
Context
Non-invasive
identification
of
individual
animals
in
wild
populations
can
occur
species
with
unique
coat
patterns
(e.g.
zebras,
giraffes,
jaguars).
However,
other
often
relies
on
capture-mark–recapture
techniques,
involving
physical
handling
animals.
Identification
koalas
(Phascolarctos
cinereus)
is
difficult
and
has
so
far
relied
mostly
invasive
methods
such
as
ear
tagging,
microchipping
and/or
collaring,
which
require
capture.
The
validation
a
non-invasive
method
to
identify
could
improve
monitoring
individuals
the
wild,
allowing
targeting
specific
disease
survival
studies,
reducing
need
capture
individuals.
Aims
This
study
describes
novel
effective
from
their
nose
markings,
specifically
using
unpigmented
pattern
determine
features
Methods
Photographs
different
Victoria
New
South
Wales
(NSW),
Australia,
were
examined
study.
Nose
traced
photographs
matched
through
visual
assessment
if
they
thought
belong
same
individual.
Differences
success
between
datasets
effect
sex
match
evaluated
statistically.
For
NSW
koalas,
lighting
conditions
photographic
angle
also
assessed.
Key
results
Overall
was
89.7%
(range
87.1–91.8%)
not
affected
by
any
variables
tested,
demonstrating
that
be
used
reliably
koalas.
Conclusions
proposed
simple,
yet
accurate
stable
over
time,
hence
it
offers
vital
tool
for
endangered
whilst
minimising
human
interference.
Implications
Pattern-based
recognition
cost-effective,
reduces
stress
animals,
potential
data
collection
allows
involvement
citizen
scientists
or
bioRxiv (Cold Spring Harbor Laboratory),
Год журнала:
2024,
Номер
unknown
Опубликована: Ноя. 11, 2024
Abstract
This
study
investigated
the
use
of
multi-sourced
photographs,
such
as
those
from
social
media,
camera
traps,
and
field
surveys,
for
individual
identification
male
Asian
elephants
to
their
corridor
use,
crop-raiding
behaviour,
occurrence
musth,
ranging
external
body
injuries
in
Thirunelli-Kudrakote
elephant
India’s
Western
Ghats.
A
total
330
images
videos
were
analysed
over
11
years
(January
2014
March
2024),
leading
27
corridor.
Eleven
individuals
observed
across
multiple
years,
demonstrating
fidelity
area.
Ten
engaging
crop-raiding,
with
four
consistently
raiding
crops
within
The
also
assessed
home
range
6
adjacent
landscapes
based
on
identification.
Overall,
shows
that
identifying
different
locations
times
provides
valuable
data
is
a
practical,
cost-effective
method
investigating
wide
questions
concerning
ecology
behaviour
elephants.
findings
underscore
potential
citizen
science
initiatives
enhance
research
elephant-range
countries.
