A new approach combining CNN, RNN, and an improved Otsu threshold method for detecting hand gestures in people with thumb finger size problems and hand tremors
Malik Kareem Kadhim,
No information about this author
Chen Soong Der,
No information about this author
Chai Phing Chen
No information about this author
et al.
AIP Advances,
Journal Year:
2025,
Volume and Issue:
15(3)
Published: March 1, 2025
The
recognition
of
hand
gestures
involves
the
application
mathematical
algorithms
to
detect
human
movements,
with
diverse
applications
in
communication
for
hearing
impaired,
human–computer
interaction,
autonomous
driving,
and
virtual
environments.
This
research
presents
a
comprehensive
approach
identifying
dynamic
gestures,
which
is
particularly
beneficial
individuals
finger
disabilities.
In
addition,
those
tremors
may
encounter
challenges
when
using
interaction
devices.
proposed
technique
enhances
sensitivity
these
devices
through
an
advanced
Otsu
segmentation
method.
It
begins
by
isolating
from
complex
backgrounds
this
sophisticated
algorithm
incorporates
motion
data
derived
RGB
video
sequences.
are
then
transformed
into
texture
contour
characteristics,
subsequently
input
hybrid
architecture
that
combines
convolutional
neural
network
(CNN)
recurrent
(RNN).
Our
findings
demonstrate
method
achieves
superior
results
existing
alternatives
can
joint
interactions
high
sensitivity.
When
comparing
traditional
our
method,
indicate
improvement
6.3%
accuracy
CNN
RNN
classifiers.
performance
novel
has
been
evaluated
compared
metrics,
yielding
significant
results.
Language: Английский
High Strength and Toughness Epoxy Nanocomposites Reinforced With Epoxy‐Terminated Hyperbranched Polymer
Hongyu Fan,
No information about this author
Tao Sun,
No information about this author
Shiji Zhang
No information about this author
et al.
Journal of Applied Polymer Science,
Journal Year:
2025,
Volume and Issue:
unknown
Published: April 17, 2025
ABSTRACT
Epoxy
composites
reinforced
and
toughened
with
an
epoxy‐terminated
hyperbranched
polymer
(ETHP)
were
successfully
synthesized.
The
ETHP
was
derived
from
a
bio‐based
epoxy
resin,
its
structure
confirmed
using
nuclear
magnetic
resonance
(NMR)
spectroscopy
Fourier
transform
infrared
(FTIR)
spectroscopy.
influence
of
content
on
the
mechanical
thermophysical
properties
systematically
investigated.
exhibited
initial
increase
followed
by
decline
increasing
loading.
7.5
wt%
ETHP/epoxy
composite
demonstrated
optimal
performance,
tensile
strength
91.7
MPa,
elongation
at
break
6.7%,
flexural
145
MPa.
Compared
to
cured
polymer,
they
represented
increases
30.6%,
34.3%,
38.1%,
respectively.
glass
transition
temperature
(Tg)
also
first
then
decreases.
A
maximum
Tg
is
obtained
when
5
wt%.
It
shifted
174.9°C
158.6°C
pure
polymer.
excellent
overall
performance
not
only
broadens
potential
applications
resins
but
provides
novel
approach
for
toughening
other
resin
systems.
Language: Английский